Test before you build.

Act as a business validation researcher. Help me define exactly what you want to test. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Rough idea: [notes]. Founder skills: [skills]. Proposed customer outcome: [outcome].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Rewrite the idea as a specific customer, situation, painful task, proposed outcome and delivery method. Produce three narrower versions rather than one broad claim. For each, list what the customer does today, the behavior that would demonstrate demand and the assumption most likely to invalidate it. Distinguish a product feature from an actual customer benefit. Do not pick a winner based on how exciting it sounds.

RETURN
A one-sentence concept, a three-option comparison table and a prioritized list of five assumptions with observable tests.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether your features solve a worthwhile problem. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Features: [list]. Customer workflow: [steps]. Claimed benefits: [claims].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map each feature to a concrete task, measurable customer outcome and current workaround. Flag features with no clear user problem. Identify whether the claimed outcome saves time, reduces cost, avoids risk or creates revenue, and specify how that outcome could be measured. Suggest a feature-free interview question for each benefit and identify one existing behavior that would challenge it. Separate convenience from an urgent need.

RETURN
A feature-to-outcome table, unsupported benefit claims and three benefit hypotheses to investigate first.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me make hidden business assumptions visible. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Business model: [how money changes hands]. Delivery process: [how it works]. Known constraints: [limits].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Break the idea into desirability, reachability, willingness to pay, delivery feasibility and financial sustainability. List at least three assumptions in each area. Rank by consequence if false, uncertainty and cost of investigation, explaining the scoring without false precision. Identify assumptions that depend on other assumptions. Choose the first three to test and propose a cheaper alternative to building software for each.

RETURN
An assumption register with IDs, rationale, evidence needed, dependencies and a sequenced first-test plan.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me choose which idea deserves a small first experiment. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Ideas: [up to three]. Relevant experience: [experience]. Resources: [assets]. Personal goals: [goals].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Define comparison criteria before evaluating the ideas: problem evidence, access to customers, delivery burden, startup cost and fit with my constraints. Use evidence-linked qualitative ratings rather than invented success probabilities. Show how rankings change under different priorities. Distinguish the best business hypothesis from the easiest idea for me to test. Give each idea a cheap experiment that could overturn its initial ranking.

RETURN
A comparison matrix, a sensitivity discussion, the biggest unknown per idea and a recommendation for the first experiment.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me assess whether you can execute the first test. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
My skills and gaps: [list]. Network: [who I can reach]. Weekly capacity: [hours]. Cash limit: [amount].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Separate skills required to validate from skills required to operate at scale. Identify tasks I can do now, learn quickly, outsource or avoid through a simpler model. Estimate effort as explicit ranges with assumptions. Surface conflicts with my schedule and resources. Propose a smaller version that tests the same customer problem. Do not infer competence or founder-market fit from ambition, degrees or interest alone.

RETURN
A capability-gap table, a minimum viable validation role description and a realistic two-week workload.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me describe the problem without pitching the solution. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Problem notes: [notes]. Customer group: [group]. Observed incidents: [examples].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Write five problem statements using the customer situation, difficult task, consequence and current workaround. Remove solution language, inflated adjectives and assumptions about emotions. Identify which statements are supported by actual incidents and which need interviews. Suggest a neutral way to ask customers whether the situation happens and what they did last time. Compare the narrowest useful statement with a broader alternative.

RETURN
Five candidate statements, an evidence check for each and one recommended research question.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me test whether your idea relies on temporary attention. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Trend: [topic]. Proposed offer: [offer]. Evidence of interest: [sources or metrics].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Identify the underlying customer task and whether it existed before the trend. Separate media attention, searches, free engagement and paid behavior. Create scenarios for attention falling sharply, stabilizing or growing; do not assign invented probabilities. List which parts of the offer retain value in each scenario. Design one test of recurring usefulness and one test of immediate purchase intent.

RETURN
A trend-versus-need table, three scenarios, weak-signal warnings and two concrete test briefs.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me design a validation plan that fits limited spare time. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Employment constraints: [hours and restrictions]. Income goal: [target]. Savings runway: [months].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Define milestones that can be reached outside working hours: problem interviews, a paid trial and repeat demand. Separate revenue from profit and cash received from commitments. Flag employment or licensing questions for verification without issuing legal conclusions. Estimate workload and opportunity costs from supplied figures. Set evidence gates for increasing hours or spending, and distinguish a test milestone from a reason to leave employment.

RETURN
A staged side-business roadmap, weekly workload, financial assumptions and evidence gates for the next commitment.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find a narrow starting point you can validate. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Full vision: [description]. Potential customers: [groups]. Proposed features: [list].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Identify the smallest customer task that carries meaningful value on its own. Produce three beachhead options with excluded features and excluded audiences. Explain what each option tests and what it cannot establish about the full vision. Choose based on reachability and existing problem evidence. Define an observable completion event and a practical offer that can be delivered manually.

RETURN
Three narrow concepts, a scope boundary for each and a single-task test specification.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me give your first experiment a clear objective. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Decision to make: [decision]. What I think is true: [belief]. Evidence so far: [notes].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Turn my notes into an experiment brief naming the customer, problem, assumption and decision at stake. Define eligible participants, recruitment source, procedure, budget, timeline and the evidence to record. Set continue, revise and stop criteria before results arrive. Explain possible false positives and false negatives. State which conclusions the experiment cannot support and assign an owner to each action only from supplied names.

RETURN
A one-page brief plus a short checklist that confirms the test could produce an inconvenient answer.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find a reachable customer group with a specific problem. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Potential segments: [list]. Existing customer access: [channels]. Problem observations: [incidents].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Split broad audiences into segments based on situation, workflow, buying trigger and ability to pay. Compare problem frequency, consequences, current spending and practical recruitment access. Avoid demographic personas with invented hobbies or feelings. Identify which differences would change the offer. Select one segment to investigate first and one adjacent segment as a comparison, explaining the evidence gaps.

RETURN
A segment comparison, a working target-customer description and recruitment criteria for each selected segment.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me identify who uses the offer and who controls purchase. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Product or service: [offer]. Organization type: [type]. Known buying process: [notes].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map users, economic buyers, budget holders, approvers, procurement staff and blockers, but label unknown roles as hypotheses. For each role, state desired outcomes, possible objections and evidence needed. Trace a proposed purchase from problem recognition to payment. Identify whether user enthusiasm can translate into budget approval and design questions to verify decision authority without collecting private organizational information.

RETURN
A buying-role matrix, a purchase-path outline and interview questions for verifying each role.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me collect evidence about actual customer behavior. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Customer segment: [segment]. Problem hypothesis: [problem]. Interview duration: [minutes].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Create a chronological interview sequence about the most recent occurrence: trigger, actions, tools, time spent, consequences and resolution. Include neutral probes when a person gives vague or generalized answers. Avoid future-intent questions such as whether they would buy an imagined product. Specify how to distinguish direct recollections from estimates and hearsay. Add a respectful way to ask for existing artifacts with permission.

RETURN
A 20-minute interview script, optional probes and a note-taking template for concrete incidents.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find the event that starts a buying journey. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Offer: [offer]. Customer notes: [notes]. Known purchase examples: [examples].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Extract triggering events from my evidence and separate ongoing annoyance from an urgent buying situation. Map what happens immediately before search, comparison and purchase. Identify timing, budget and authority changes that may matter. Generate alternative trigger hypotheses only when labeled as such. Design interview questions and a targeted recruitment description for people who recently experienced each trigger.

RETURN
A trigger map, evidence strength for each trigger and a test plan for the highest-priority buying situation.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me understand the behavior your offer must replace. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Current solutions: [tools or processes]. Interview notes: [notes]. Proposed offer: [offer].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Describe each workaround step by step, including free tools, manual work, staff time and doing nothing. Identify benefits the current approach provides despite its weaknesses. Estimate costs only from supplied or sourced figures and show formulas. Explain switching friction, trust and habits that could make a better-looking solution unattractive. Propose a test where customers try one replaced step rather than abandon their whole workflow.

RETURN
A workaround map, replacement barriers and a single-step comparison experiment.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me distinguish irritation from a problem people act on. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Reported complaints: [notes]. Actual actions taken: [actions]. Costs or consequences: [figures].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Assess frequency, severity, urgency and effort already spent solving the problem. Separate stated frustration from observed action. Use anchored qualitative categories with definitions and missing evidence, rather than a fabricated numeric pain score. Look for counterexamples: customers who experience the issue but comfortably ignore it. Create questions that test whether consequences matter enough to justify time, payment or switching.

RETURN
A problem-strength matrix, contradictory evidence and a concrete test of customer action.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me avoid testing with customers who cannot use or buy it. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Proposed target group: [group]. Delivery requirements: [requirements]. Offer limits: [limits].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Define observable exclusion criteria such as incompatible workflow, lack of authority, inaccessible location or a need the offer cannot meet. Distinguish poor fit from a customer who raises useful criticism. Explain how each exclusion changes recruitment and marketing claims. Create screening questions that avoid sensitive personal details. Describe a nearby segment that appears attractive but would create misleading validation results.

RETURN
An exclusion list, a respectful screening form and a warning about likely sample contamination.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me learn why existing solutions lost users. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Alternative products: [names]. Churn interviews or reviews: [text]. Proposed difference: [difference].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Group abandonment reasons into unmet outcomes, cost, setup friction, trust, changed needs and lack of recurring use. Keep quoted evidence tied to its source and do not invent motives. Separate failures specific to one product from structural problems in the category. Design follow-up questions about what people used afterward and whether the original need persisted. Test whether my proposed difference addresses a cause or merely a symptom.

RETURN
An abandonment-pattern table, category risks and five questions for former users.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find out whether customers face a meaningful consequence. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Workflow details: [steps]. Frequency: [times per period]. Measured costs: [figures].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Build a cost-of-inaction model using time, errors, lost opportunities and operational risk, but quantify only supported variables. Keep monetary costs separate from unpriced consequences. Show low, base and high assumptions with no double counting. Ask which consequences customers themselves consider significant. Design a way to verify one key variable through records or timed observation rather than asking for vague estimates.

RETURN
A transparent calculation, an assumptions table and a plan to verify the largest uncertain cost.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me turn research notes into a useful working customer description. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Interview notes: [anonymized notes]. Sample details: [how participants were recruited].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Create a customer profile based on recurring situations, tasks, triggers, alternatives, buying roles and constraints. Attach every substantive attribute to evidence or label it untested. Do not invent demographic detail, psychological traits or quotes. Identify disagreements between participants and avoid averaging incompatible segments. Include the profile scope, who it excludes and what additional evidence could change it.

RETURN
A one-page working customer profile, source-linked evidence and the next three research gaps.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me plan a customer interview sample with clear eligibility. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Research question: [question]. Channels available: [channels]. Desired sample: [number].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Define participant eligibility around real recent behavior, not enthusiasm for startups. Propose recruitment sources, a neutral invitation and a screener. Include a mix of current users, alternative users and people who chose not to solve the problem where relevant. Flag convenience sampling and incentive bias. Explain what a small qualitative sample can reveal and what it cannot establish about market prevalence.

RETURN
A recruitment plan, short invitation, five-question screener and sampling limitations.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me learn about the problem without selling your idea. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Problem: [problem]. Participant type: [type]. Duration: [minutes].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Write an opening consent statement and a script focused on recent events, workflow, alternatives and costs. Order questions from broad context to concrete details. Mark optional probes and questions to avoid because they introduce the solution or invite compliments. Include a way to ask about purchase authority and existing budgets without pressuring disclosure. End with a summary check that lets the participant correct my interpretation.

RETURN
A timed interview script, interviewer instructions and a structured notes sheet.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me repair research questions that push people toward agreement. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Draft questions: [paste]. Intended decisions: [decisions].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Audit each question for leading wording, double-barreled phrasing, vague timeframes, forced assumptions and social desirability bias. Rewrite questions with clear periods and balanced answer options including not applicable. Distinguish behavior measures from opinions and hypothetical buying intent. Explain which question can inform which decision. Preserve my research goals while cutting questions that gather interesting but unusable information.

RETURN
A before-and-after table, a revised survey and a decision-to-question mapping.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me measure reported experiences without pretending they prove demand. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Audience: [audience]. Problem: [problem]. Distribution channel: [channel].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Build a survey taking about three minutes that asks about recent problem occurrence, frequency, consequences, current solutions and existing spending. Include screening, clear timeframes, balanced choices and one optional open question. Do not include a purchase-intent question as the sole demand measure. Explain distribution bias and nonresponse. Specify a separate behavioral test for the strongest hypothesis that survey results might generate.

RETURN
A publishable survey, coding instructions and an interpretation guide with limitations.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me investigate real purchasing constraints. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Offer concept: [concept]. Target buyer: [buyer]. Current alternatives and costs: [notes].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Create questions about past spending, budget source, approval limits and tradeoffs before showing an offer. If proposing prices, label reactions as stated preferences and separate them from purchases. Suggest a realistic next-step commitment such as a properly disclosed paid trial. Avoid manipulative urgency and questions designed to obtain a yes. Identify when the person lacks authority to provide meaningful pricing evidence.

RETURN
A pricing interview script, commitment options and an evidence hierarchy from opinion to paid behavior.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find patterns and contradictions in qualitative evidence. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Notes or transcripts: [paste]. Research objective: [objective]. Sample and recruitment: [details].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Code statements into problem incidents, workaround, consequence, trigger, buyer constraint and counterevidence. Preserve short exact quotes only from supplied text and identify participant IDs. Count participants mentioning a theme, not repeated mentions from one person. Distinguish observed behavior from interpretation. Look for divergent segments and contradictions. State whether the notes are sufficient for conclusions and identify where follow-up questions would improve confidence.

RETURN
A theme table with source IDs, contradictory cases, limitations and a prioritized follow-up list.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me interpret encouraging feedback without calling it validation. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Feedback: [messages or notes]. Actions requested: [requests]. Actions completed: [records].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Classify responses as praise, stated interest, effortful participation, contact permission, trial use, payment or repeat purchase. Explain the limits of each category in this context. Identify missing denominators and selection effects. Rewrite any unsupported conclusions about demand. Propose a proportionate next ask that lets people decline easily and produces a more meaningful signal than another compliment.

RETURN
An evidence classification table, corrected conclusions and three ethical next-step requests.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether conflicting feedback reveals distinct segments. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Conflicting notes: [paste]. Participant context: [roles and workflows].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Separate contradictions caused by different roles, situations, experience levels, problem frequency or offer interpretations. Identify which explanations have evidence and which are speculative. Create follow-up questions to distinguish them. Avoid selecting only the comments that support my idea. Show whether the evidence suggests separate segments, unclear communication or a weak overall problem. Propose an experiment that could challenge the favored explanation.

RETURN
A contradiction map, segmentation hypotheses and a focused follow-up research plan.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether people understand the offer before judging it. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Concept description: [copy]. Intended audience: [audience]. Claims: [claims].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Create a test where participants first explain the offer in their own words. Ask what they think it does, who it is for, what result it promises and what is unclear before asking preference questions. Keep clarifications until after the initial response. Record misunderstanding separately from dislike. Include alternative descriptions that vary one element and a plan to compare comprehension without treating the interview as a sales pitch.

RETURN
A comprehension script, coding rubric and a comparison plan for two descriptions.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me learn how the task happens in practice. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Task: [task]. Setting: [setting]. Participant access: [access].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Design a consent-based observation plan recording steps, handoffs, delays, errors and tools. Define what to observe without collecting confidential content. Separate observer notes from participant explanations. Include a way to record time and interruptions consistently. Explain how observation may alter behavior and how one setting may limit transferability. Propose a follow-up interview about discrepancies between stated workflow and observed actions.

RETURN
An observation protocol, event log template and questions for checking interpretation with participants.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me use public discussions to generate testable problem hypotheses. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Category: [category]. Customer type: [type]. Keywords: [phrases]. Public links: [optional].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
If browsing is available, search relevant public Reddit discussions about specific frustrations, workarounds, recommendations and abandoned solutions. Otherwise give exact search queries and ask me to paste posts. Record dates, community, context and source URLs. Separate frequent mentions in the sampled posts from prevalence in the wider market. Look for contradictions and promotional content. Do not infer paid demand from upvotes or fabricate threads.

RETURN
A source-linked problem table, recurring wording, sample limitations and three hypotheses to test with customers.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me use question patterns to identify gaps worth investigating. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Topic: [topic]. Audience: [audience]. Accessible question links or text: [paste].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Analyze accessible Quora questions for recurring tasks, confusion, alternatives and attempts to solve the problem. If access is blocked, say so and provide manual search queries; never invent questions or answers. Separate unanswered questions from proof of a commercial opportunity. Check dates and contextual differences. Identify information needs that a free explanation could satisfy and needs that might justify a product, then design a behavioral test for the latter.

RETURN
A question-pattern table, source limitations and a plan to investigate commercial intent.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find problems competitors repeatedly fail to solve. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Review text or URLs: [sources]. Category: [category]. Date range: [range].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Group reviews by specific failure, affected customer, situation and consequence. Track source, date and product version where available. Separate delivery failures, support complaints and unrealistic expectations from missing product capabilities. Include positive reviews as counterevidence and flag authenticity uncertainty. Do not assume reviewers represent all users. Suggest one improvement hypothesis per pattern and a question to verify whether non-reviewers share the problem.

RETURN
A review-coding table, opportunity hypotheses and a plan to validate them beyond review sites.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me investigate what people mean when they search. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Keywords: [list]. Search data: [source and date]. Offer: [offer].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Classify queries as informational, comparison, transactional or ambiguous and explain the classification. Verify search volumes only from provided or accessible sources. Map queries to likely tasks without treating volume as customers. Identify queries where a free answer satisfies the need and queries where a paid offer may fit. Design a landing-page or offer test with a meaningful conversion event and clear eligibility criteria.

RETURN
A keyword-intent map, evidence gaps and a small search-demand experiment.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me avoid mistaking a short spike for consistent demand. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Demand records: [data]. Dates and geography: [details]. Relevant events: [events].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Inspect the supplied time series for seasonality, one-off spikes, reporting changes and promotion effects. If current external data is needed, use sources with dates or provide a collection plan. Compare equivalent periods rather than adjacent months alone. State when the available history is too short. Design tests in both peak and ordinary periods and explain the operating implications of seasonal revenue.

RETURN
A seasonality assessment, alternative explanations and a timing plan for validation.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me model who you could realistically reach and serve. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Geography: [area]. Eligible customer criteria: [criteria]. Channel capacity: [figures]. Delivery capacity: [limits].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Build an estimate starting with verified eligible customers rather than broad industry revenue. Show each filter, denominator, source and assumption. Separate total eligible customers, reachable customers and capacity-limited sales. Use ranges and sensitivity rather than a single precise market-size number. Do not multiply an arbitrary market share by a large market. Identify the input that most affects the estimate and how to verify it.

RETURN
A bottom-up worksheet, source list, sensitivity cases and the next data collection action.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me judge what your available evidence actually supports. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Signals: [searches, interviews, signups, trials, purchases]. Collection method: [details].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Evaluate each signal for proximity to actual value, effort required, buyer authority, incentive bias, recency and repeatability. Keep counts tied to denominators and customer eligibility. Explain why money, time or usage can be meaningful but still context-dependent. Identify signals generated by the same people so they are not double counted. Propose the smallest next test that improves the weakest important assumption.

RETURN
An evidence ladder, signal-quality table and a recommendation with explicit limitations.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find evidence of problem-driven action. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Problem: [problem]. Customer segment: [segment]. Research notes: [notes].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Identify behaviors showing active search: asking peers, comparing tools, requesting quotes, budgeting or trying workarounds. Distinguish passive exposure to advertising from self-initiated search. Create research queries and interview questions about recent searches. Consider people who stopped searching or chose to do nothing. Define what evidence would support a high-intent segment and what would challenge it, without assuming all problem sufferers seek a solution.

RETURN
A search-behavior checklist, interview prompts and an actionable segment hypothesis.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether online engagement can support a paid offer. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Community: [name]. Engagement examples: [posts]. Offer and price: [details].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Audit who engaged, whether they match the buyer, what they reacted to and whether incentives or novelty drove interest. Separate upvotes, views and comments from purchases or useful trials. Look for critical comments and missing voices. Suggest a disclosed offer experiment that complies with community rules and does not spam members. Define an eligible conversion denominator and alternative explanations if the test gets no sales.

RETURN
An engagement audit, an ethical offer-test brief and an interpretation matrix.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me evaluate research before using it in a business decision. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Sources: [links or excerpts]. Decision: [decision]. Claimed facts: [claims].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Assess author, methodology, date, geography, sample, commercial incentives and relevance to my customer. Trace important statistics to their original source where accessible. Flag summaries without methods, stale estimates and circular citations. Separate anecdotes from descriptive data and causal claims. Suggest replacements for weak sources or practical primary research when no useful data exists. Avoid treating a famous publisher as proof of methodological quality.

RETURN
A source-audit table, corrected claims and a research gap list prioritized by decision impact.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me understand all the alternatives customers can choose. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Offer: [offer]. Known competitors: [names]. Customer task: [task].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map direct competitors, adjacent products, freelancers, internal staff, free tools and doing nothing. Compare customer outcomes, buying situation, price structure, delivery burden and switching cost. Use current sources for claims where possible and mark unknowns. Explain why the absence of direct competitors could reflect either opportunity or weak demand. Select three alternatives for closer customer research rather than declaring the market empty.

RETURN
An alternatives matrix, source links and a shortlist for the next investigation.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me compare what buyers actually get for the advertised price. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Competitors and links: [list]. Customer use case: [use case]. Currency and region: [details].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Read accessible current official pricing pages or use my pasted data. Record billing period, currency, taxes if specified, usage limits, minimum term, add-ons and setup costs. Distinguish advertised prices from quote-based or unavailable prices. Calculate comparable costs only for a defined usage scenario. Do not guess discounts or private enterprise quotes. Explain how price structure affects my proposed test rather than recommending automatic undercutting.

RETURN
A dated price-and-terms table, normalized scenario costs and unresolved comparison questions.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me validate differentiation through customer tradeoffs. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Claimed difference: [difference]. Target customer: [customer]. Alternatives: [alternatives].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Translate the difference into a measurable customer outcome and identify the tradeoff it requires. Separate distinctive marketing from useful functional value. Create a neutral comparison where customers choose between realistic offers without brand cues where feasible. Ask why they chose and what they would give up. Specify which result would suggest the difference is irrelevant, valuable only to a niche or insufficient to overcome switching costs.

RETURN
A differentiation hypothesis, comparison-test materials and an interpretation plan.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me turn your evidence into clear alternative promises. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Customer evidence: [notes]. Offer: [offer]. Known limitations: [limits].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Write five value propositions naming the customer, concrete problem, expected outcome and mechanism. Avoid guaranteed results, unverifiable superlatives and invented proof. Make each version test a genuinely different hypothesis such as time savings, reduced errors or simpler setup. Define one comprehension measure and one behavioral response for comparison. Explain which claim needs evidence before appearing publicly and how to collect that evidence.

RETURN
Five propositions, the underlying hypothesis for each and a practical message-test brief.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether a low price solves the real buying barrier. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Competitor costs: [verified figures]. My costs: [figures]. Proposed price: [amount].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Analyze whether price is a demonstrated objection or my assumption. Include perceived quality, trust, switching burden, support and delivery economics. Model contribution margin at proposed prices using supplied costs. Design a test that distinguishes price sensitivity from lack of value without changing several variables simultaneously. Explore a narrower scope or different payment structure if a discount makes delivery unsustainable.

RETURN
A pricing-strategy critique, margin scenarios and a test that can distinguish competing explanations.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me locate a specific mismatch between alternatives and customers. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Category: [category]. Review or interview evidence: [notes]. Candidate niches: [list].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Identify mismatches involving workflow, language, location, business size, accessibility or delivery format only when evidence supports them. Assess whether the niche is reachable and can pay rather than merely distinct. Include current workarounds and reasons incumbents may avoid the segment. Propose a narrow offer and an experiment to test whether the mismatch causes enough action to sustain a business.

RETURN
A niche-opportunity table, evidence gaps and a first-offer experiment for one candidate.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether a paid offer earns its place. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Free alternatives: [tools]. Proposed price: [price]. Outcome: [outcome].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map what the free approach already accomplishes and where it costs time, skill, coordination or risk. Do not assume convenience justifies payment. Design a realistic task comparison measuring completion, quality and effort. Identify customer segments for whom the free solution is enough. Create a paid trial offer only for the remaining segment and specify what usage or purchase evidence would justify further investment.

RETURN
A free-versus-paid comparison, eligible segment and task-based validation protocol.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me understand one competitor beyond its homepage claims. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Competitor URL or materials: [sources]. My target customer: [customer].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Review accessible positioning, onboarding, offer structure, pricing, trust evidence and public feedback. Keep observed details separate from assumptions about internal performance. Do not infer revenue, ad spend or conversion rates from design quality. Identify which choices fit its audience and which may not transfer to mine. Propose customer questions to test observed friction and avoid recommending a copy of its brand or proprietary content.

RETURN
An evidence-linked teardown, transferable hypotheses and a validation checklist.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find whether one offer needs different messages. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Segments: [A and B]. Offer: [offer]. Research evidence: [notes].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Create one plain-language proposition for each segment based on its task, trigger and buying criteria. Keep the underlying offer stable so the test compares relevance rather than different products. Define recruitment, equivalent exposure and response measures. Record misunderstanding separately from lack of interest. Explain how unequal channel quality could confound results and what additional evidence is needed before choosing a primary segment.

RETURN
Two message briefs, a segment comparison protocol and a decision table for likely outcomes.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me identify the barrier between liking and adopting your offer. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Current solution: [solution]. Proposed offer: [offer]. Setup and migration requirements: [details].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map data migration, training, habits, contracts, trust, compatibility and approval friction. Ask about past switches and what triggered them. Separate one-time transition costs from recurring value. Design a small trial that reduces irreversible commitment while retaining a meaningful real task. Specify how to observe adoption rather than just account creation. Identify the minimum outcome required to make a switch rational for this customer.

RETURN
A switching-cost map, interview questions and a low-friction adoption experiment.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me create a small offer that tests willingness to pay. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Proposed outcome: [outcome]. Delivery capacity: [limits]. Price hypothesis: [amount].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Define a bounded pilot with eligible customers, deliverables, timeline, price, responsibilities and success measures. State what is unfinished and what the customer will actually receive. Avoid promises that depend on unverified demand or third-party results. Include a fair decline path and any refund terms as proposals requiring local verification. Explain what pilot payment, usage and repeat interest would each demonstrate.

RETURN
A pilot offer sheet, invitation draft and a results log separating payment from delivered value.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me compare payment structures against how value is delivered. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Offer: [offer]. Usage pattern: [pattern]. Cost drivers: [costs]. Buyer preferences: [evidence].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Compare one-time purchase, subscription, per-use, retainer and outcome-linked pricing where relevant. Explain which fit recurring value and which shift risk or create complexity. Model cash timing and contribution margin with supplied inputs. Identify the strongest assumption for each model, such as repeat need or measurable outcomes. Recommend one simple initial experiment rather than a complex pricing system.

RETURN
A pricing-model comparison, a recommended test and the assumptions that could reverse that recommendation.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me create a price hypothesis without pretending it is optimized. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Delivery costs: [figures]. Comparable prices: [sources]. Customer outcome: [outcome].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Calculate a cost floor with variable delivery, support, payment fees and refunds where supplied. Separate owner compensation from profit. Compare relevant alternatives without assuming their price is affordable to my audience. Propose a small set of price hypotheses with rationale and a real offer test. Clarify that the result estimates response under these conditions, not a universal optimal price.

RETURN
A transparent price worksheet, candidate prices and a test plan with decision thresholds.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether customers need recurring value. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Recurring offer: [offer]. Usage cycle: [cycle]. One-off alternative: [alternative].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Identify what value renews each period and what may be exhausted after one use. Design a manually delivered sequence that observes repeat use and renewal at a disclosed price. Compare recurring and one-off options only with clear scope. Track cancellations and reasons rather than relying on claimed renewal intent. Explain how introductory discounts, incentives and founder attention could inflate early retention.

RETURN
A recurring-value hypothesis, a manual renewal pilot and a retention evidence checklist.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me test whether a specific audience values the outcome. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Guide topic: [topic]. Target reader: [reader]. Proposed price: [amount]. Draft outline: [outline].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Separate demand for the topic from demand for this paid format. Assess what free content already solves. Define a narrow outcome, sample chapter and small paid workshop or disclosed preorder test where fulfillment is feasible. Ask customers about past purchases and unfinished guides. Plan to measure sample use, purchase and completion separately. Avoid promising income or results the guide cannot guarantee.

RETURN
A digital-guide validation brief, sample outline, honest offer copy and a decision framework.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether customers will buy a bounded service. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Service skills: [skills]. Customer task: [task]. Proposed deliverables: [list].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Turn the service into a clearly scoped package with one outcome, exclusions, delivery time and client inputs. Estimate fulfillment hours using explicit assumptions and include revisions and support. Design a first-client pilot with a real price and measurable delivery outcome. Distinguish willingness to pay from the economics of delivering. Identify custom requests that would invalidate the standardized package assumption.

RETURN
A service-package sheet, delivery-time model, pilot proposal and scope-risk checklist.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me investigate demand before committing to stock. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Product: [product]. Supplier information: [details]. Unit and shipping costs: [figures].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Compare samples, small batches, demonstrations and clearly disclosed preorder options. Identify product-quality, shipping, return and fulfillment assumptions independently from customer interest. Use supplied cost data and mark missing duties or compliance checks for verification. Design a realistic offer with delivery conditions and availability accurately stated. Include a stop-loss budget and a method to record purchase objections without buying speculative inventory.

RETURN
A low-inventory test plan, fulfillment questions and a unit-cost worksheet.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me learn whether a business can approve a real purchase. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Business buyer: [role]. Offer: [offer]. Typical buying process: [notes].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Create a discovery-to-pilot sequence that verifies problem ownership, economic buyer, budget source, security or procurement needs and timeline. Separate user enthusiasm from approval authority. Draft a pilot proposal with scope, outcome measurement and a realistic decision date. Identify commitments that are revocable or nonbinding. Propose a way to verify purchasing progress without overstating a verbal yes or letter of intent.

RETURN
A B2B validation sequence, buyer questions and a pilot proposal outline.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me determine whether customers want the combined offer. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Bundle components: [list]. Standalone alternatives: [prices]. Customer task: [task].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map which components address one coherent outcome and which merely add volume. Design a choice test comparing a core offer with a clearly priced bundle. Keep differences explicit and avoid fabricated savings. Ask what customers would remove and why. Measure purchase behavior and use of each component separately. Consider whether a popular bundle masks an unneeded feature or creates costly fulfillment obligations.

RETURN
A bundle comparison, test materials and a component-use measurement plan.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find out what a too-expensive response means. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Objections: [exact anonymized wording]. Offer and price: [details]. Buyer context: [notes].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Classify objections as affordability, unclear outcome, low urgency, weak trust, comparison mismatch, timing or bargaining, but tie interpretations to evidence. Write neutral follow-up questions rather than pressure tactics. Suggest a narrower scope, payment structure or proof test where appropriate. Calculate the effect of discounts on contribution margin. State which evidence would justify changing price and which would suggest changing the offer or target customer.

RETURN
An objection-analysis table, follow-up questions and a decision guide for offer changes.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me learn about your riskiest assumption in one week. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Assumption: [assumption]. Available channels: [channels]. Daily time: [hours].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Build a seven-day sequence beginning with evidence review and ending with a decision. Include recruitment, a real customer interaction, an offer or task test and a review. Define the key assumption, measurement and thresholds before launch. Keep the schedule inside my budget and hours, with dependencies and a fallback for slow recruitment. Explain what this sprint cannot prove about long-term demand or scalability.

RETURN
A daily plan, experiment brief, evidence log and continue/revise/stop review template.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me measure behavior beyond visits and likes. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Offer: [offer]. Traffic source: [source]. Page draft: [copy]. Target action: [action].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Define one audience, one promise and one appropriate action such as booking a qualified call or applying for a disclosed pilot. Draft clear page copy explaining availability. Set up a funnel from eligible visitors to completed action with denominators. Explain confounds from poor traffic, unclear messaging and technical failures. Propose thresholds grounded in business needs rather than invented industry benchmarks.

RETURN
A landing-page brief, copy draft, event map and results interpretation matrix.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me assess signup quality and subsequent participation. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Offer: [offer]. Audience: [audience]. Follow-up capacity: [capacity].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Design a waitlist with clear expectations, minimal data collection and one optional qualification question. Plan a follow-up invitation to a useful trial or research task. Track eligible signups, response, attendance, use and payment separately. Avoid artificial scarcity and fake launch dates. Explain why low-friction signups alone cannot establish paid demand and how incentives may distort the sample.

RETURN
Waitlist copy, follow-up sequence and a funnel worksheet from signup to meaningful commitment.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me deliver the outcome manually before automating it. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Customer outcome: [outcome]. Manual tools: [tools]. Capacity: [hours].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map how I can deliver the promised outcome by hand with clear customer knowledge of the pilot format. Define onboarding, delivery steps, quality checks and feedback collection. Log work time, exceptions and support. Separate evidence of customer value from evidence that automation will work. Identify which steps could later be automated and which require human judgment. Set limits preventing the pilot from becoming unrestricted bespoke work.

RETURN
A concierge-pilot operating plan, time log and evidence gates before automation.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find usability problems without calling them market validation. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Prototype: [description or link]. Task: [task]. Participants: [criteria].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Create a test where participants attempt one realistic task with minimal coaching. Define completion, errors, time and confidence measures. Ask comprehension questions before giving explanations. Separate usability friction from lack of need and state that task completion does not prove willingness to pay. Include accessibility and device considerations appropriate to the audience. Create a severity rubric and a retest plan focused on the most important barriers.

RETURN
A task-based test script, observation sheet and prioritized issue-report format.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me test response to an offer that is not fully built. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Unbuilt offer: [offer]. Current availability: [status]. Intended CTA: [action].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Design an offer test that accurately states what exists and what happens after the action. Avoid fake checkout confirmations, fabricated stock or undisclosed unavailable features. Choose an action proportionate to readiness, such as a pilot application. Explain which assumption it tests and how lack of fulfillment affects interpretation. Define event tracking, budget cap and a debrief or follow-up plan for participants.

RETURN
Honest smoke-test copy, a user flow, tracking requirements and decision criteria.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me investigate demand where you can actually serve customers. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Local offer: [offer]. Service area: [area]. Capacity: [limits].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Define a tightly bounded location and eligible customer group. Compare interviews, a small demonstration, local listings and a limited paid pilot. Account for travel, scheduling, local alternatives and repeat need. Identify permits or location-specific requirements for verification before fulfillment. Track inquiries, qualified bookings, completion and repeat requests separately. Explain why one neighborhood may not generalize to an entire city.

RETURN
A neighborhood test plan, outreach materials and a location-specific operating checklist.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me reduce the uncertainty in a two-sided business. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Supply side: [who]. Demand side: [who]. Transaction: [what happens].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Identify which side faces the more urgent problem and the minimum useful transaction. Map matching constraints, trust, quality and payment expectations. Design a manually coordinated pilot in one narrow category or geography. Measure match availability, accepted introductions, completed transactions and repeat use on both sides. Do not treat supplier signup as buyer demand. Identify subsidies and founder effort that could mask poor marketplace economics.

RETURN
A two-sided assumption map, manual matching protocol and liquidity evidence checklist.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me avoid experiments that are easy but uninformative. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Assumption: [assumption]. Candidate tests: [list]. Budget and time limits: [limits].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Compare tests by evidence relevance, participant fit, cost, duration, confounds and reversibility. Explain which can falsify the assumption and which merely produce engagement. Include interviewing, task observation, manual delivery and paid offers where appropriate. Pick the cheapest test that reaches the necessary evidence level, not the absolute cheapest activity. Define what follow-up would be needed if the result is ambiguous.

RETURN
A test-selection matrix, recommended experiment and fallback plan for inconclusive results.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me prevent moving the goalposts after results arrive. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Experiment: [procedure]. Business requirements: [figures]. Eligible audience: [criteria].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Define the primary metric, numerator, denominator, observation window and exclusions. Set continue, revise and stop criteria using my economics and risk tolerance rather than borrowed conversion benchmarks. Identify guardrails such as delivery cost or dissatisfaction. Specify how missing data and technical failures will be handled. Record alternative explanations and what additional test would distinguish them. Keep criteria provisional if critical inputs are unavailable.

RETURN
A preregistered decision sheet, metric definitions and an ambiguity-resolution plan.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me choose a realistic recruitment or acquisition channel. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Target customer: [criteria]. Channels available: [list]. Budget: [amount].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map where eligible customers already look for help, compare options or discuss the task. Separate channels I can access now from aspirational channels. Compare fit, cost, time, trust requirements and measurement. Use source verification for current platform rules where needed. Choose one initial channel and define a small action that can produce identifiable eligible responses. Avoid estimating conversions without relevant evidence.

RETURN
A channel-access matrix, first-channel recommendation and a measured first-week plan.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me test an offer with a controlled advertising budget. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Platform: [platform]. Offer and price: [details]. Budget cap: [amount].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Define the test assumption, eligible audience, one offer and a meaningful conversion event. Draft two distinct message hypotheses and a landing-page requirement checklist. Explain targeting limitations and differences between clicks, leads and customers. Use supplied economics to calculate acceptable acquisition cost and stop-loss rules. Include tracking validation, attribution limits and a procedure for low traffic or zero conversions. Do not invent platform benchmarks.

RETURN
An ad-test brief, two creative drafts, budget allocation and a decision table.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me recruit participants without biasing their answers. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Customer type: [type]. Research question: [question]. Recruitment setting: [setting].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Write three concise invitations tailored to the context, accurately stating the purpose, time requirement and any incentive. Describe the task area without pitching my solution or claiming a problem everyone has. Include eligibility, an easy decline and a clear scheduling next step. Explain how recruiting only enthusiastic supporters could bias results. Provide a response log that records source and eligibility without unnecessary personal data.

RETURN
Three invitation drafts, a neutral follow-up and a recruitment-quality checklist.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether another business can bring suitable customers. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Potential partners: [types or names]. Offer: [offer]. Customer overlap: [evidence].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map what each partner gains, what they must do and why their customers would care. Draft a small pilot proposal with accurate claims and clear responsibilities. Define referral qualification, attribution, delivery capacity and any proposed commercial terms. Separate a friendly agreement from actual referred buyers. Include a fallback if the partner supplies poor-fit leads and a test of whether the arrangement works without founder handholding.

RETURN
A partner-fit matrix, pilot proposal and referral-results worksheet.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me test whether satisfied users actually introduce others. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Offer: [offer]. Existing users: [count and context]. Referral process: [proposal].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Identify when customers experience enough value to make a referral and what friction prevents it. Design a modest, clearly disclosed ask after a meaningful outcome. Separate referrals offered, introductions completed, eligible leads and purchases. If incentives are proposed, explain their cost and potential bias. Avoid treating intention to recommend as a referral. Define the evidence needed before building a referral system.

RETURN
A referral experiment, request copy, tracking funnel and interpretation guide.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me determine what an early ad test says about affordability. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Spend: [amount]. Funnel counts: [counts]. Price and variable costs: [figures].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Calculate cost per eligible lead and paid customer using clear denominators and a stated attribution window. Include refunds and delayed purchases where known. Separate observed acquisition cost from assumptions about future improvement or repeat revenue. Model contribution after acquisition for a one-off sale and only include repeat purchases supported by data. Explain sample uncertainty and identify whether traffic quality, offer or economics need the next test.

RETURN
A transparent acquisition worksheet, break-even analysis and a spend/stop/retest recommendation.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me compare acquisition sources under realistic conditions. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Channels: [A and B]. Offer: [offer]. Resources per channel: [limits].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Define comparable customer eligibility, offer, measurement period and conversion event. Track monetary spend and founder time separately. Explain differences in intent, reach, lag and attribution that prevent a simple winner declaration. Keep channel-specific creative appropriate while preserving the same core offer. Propose minimum informative observations and an inconclusive outcome. State what result would justify a larger follow-up test in one channel.

RETURN
A channel comparison protocol, shared metrics and an interpretation table.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me use useful content to investigate customer problems. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Audience: [audience]. Problem hypotheses: [list]. Distribution: [channels].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Design three pieces of practical content, each addressing a different problem hypothesis. Define a relevant next action such as trying a sample or applying for a pilot rather than maximizing views. Keep availability and business affiliation clear. Track eligible responses, follow-through and stated tasks. Explain why successful free content does not prove paid-product demand and propose the next offer test if one topic attracts suitable participants.

RETURN
Three content briefs, CTA drafts and a response-coding plan.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me measure whether nearby customers request useful help. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Local service: [service]. Listing or page: [copy]. Coverage: [area].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Review the listing for clear location, scope, availability and contact expectations. Define inquiry qualification and a realistic booking step. Track discovery source, qualified inquiry, quote acceptance, completed job and repeat request. Consider whether trust, response time or scheduling cause drop-off. Verify platform and local requirements when applicable. Do not create fake reviews, false addresses or availability claims as part of the test.

RETURN
A listing-test brief, improved copy and a local inquiry funnel log.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me learn from nonbuyers without assuming the reason. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Lead records: [anonymized records]. Offer: [details]. Known drop-off stage: [stage].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Separate ineligible leads, undecided prospects, rejected offers and failed transactions. Create neutral follow-up questions about the actual decision, alternatives chosen and unresolved concerns. Avoid presuming price was the issue or interpreting silence as rejection. Group evidence by source and stage. Recommend a focused change and a way to test it without simultaneously changing audience, price and offer.

RETURN
A nonbuyer interview script, drop-off diagnosis and one controlled follow-up experiment.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether one delivered sale can make economic sense. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Price: [amount]. Variable costs: [list]. Delivery and support time: [hours]. Fees and refunds: [figures].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Compute net collected revenue, variable cost, contribution per sale and contribution after acquisition from supplied inputs. Include owner time separately and show both cash and compensated-labor views. Distinguish fixed costs from variable costs. Use low/base/high cases for uncertain inputs and state missing costs. Identify the assumption that most affects viability and design a measurement during the pilot to replace it.

RETURN
A formula-based worksheet, sensitivity cases and a pilot cost-measurement checklist.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find the sales volume required by your costs. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Fixed monthly costs: [figures]. Price: [price]. Variable cost: [cost]. Capacity: [units or hours].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Calculate break-even units using contribution per unit and show what happens if contribution is zero or negative. Include owner compensation as a separate scenario. Compare required volume with realistic delivery capacity and verified channel reach. Do not turn required sales into a forecast. Show sensitivity to price, refunds and acquisition cost, and identify which inputs need direct pilot measurement.

RETURN
A break-even table, capacity comparison and a list of feasibility assumptions to test.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me test whether fulfillment fits your available hours. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Delivery steps: [steps]. Time records: [data]. Weekly capacity: [hours].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map preparation, customer communication, delivery, revisions, support and administration. Use measured times where available and label estimates. Model typical and difficult cases, including interruptions and rework. Calculate capacity without assuming every hour is productive. Identify bottlenecks and tasks that scale poorly. Design a time log for the next few deliveries and an offer boundary that keeps the pilot inside capacity.

RETURN
A fulfillment-time model, weekly capacity estimate and a delivery log template.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check hidden costs that early revenue can conceal. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Offer: [offer]. Refund policy proposal: [terms]. Support records or estimates: [data].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map likely support requests, disputes, quality failures and refund causes as hypotheses. Calculate scenarios using explicit cost and rate assumptions rather than industry averages. Separate one-time onboarding issues from recurring delivery defects. Identify what to log during pilots and which promises may increase burden. Propose clear scope and expectations that reduce preventable problems without obstructing legitimate complaints.

RETURN
A support/refund risk model, logging plan and revised expectation-setting checklist.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me verify that a physical offer can be fulfilled consistently. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Supplier details: [details]. Sample results: [results]. Lead times and minimum order: [figures].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Create a supplier verification plan covering samples, quality variation, lead time, replacement handling, capacity and documented terms. Separate supplier claims from verified deliveries. Model how delays or defects affect the pilot offer and cash needs. Identify relevant safety or import checks for qualified verification. Suggest a small order test with acceptance criteria and a fallback rather than buying inventory based only on a catalog.

RETURN
A supplier test checklist, acceptance criteria and a fulfillment-risk table.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether the tools can deliver a reliable customer outcome. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Tools: [names and plans]. Workflow: [steps]. Quality requirements: [criteria].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Identify dependencies on pricing, usage limits, access, exports, reliability and human review. Verify current official documentation when accessible and otherwise list checks. Test representative and difficult tasks with explicit quality criteria. Include failure handling and data minimization appropriate to the use case. Separate a successful demo from reliable repeated operation and calculate per-delivery tool costs using known inputs.

RETURN
A dependency map, quality-test protocol and a fallback delivery plan.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me keep investigation spending proportionate to what you can learn. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Total cash limit: [amount]. Available hours: [hours]. Candidate activities: [list].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Allocate the budget by the highest-risk assumptions and evidence value of each activity. Separate reusable assets from test-specific spending and include recruitment, fulfillment and contingency where applicable. Define a stop-loss cap and an approval point before each larger commitment. Avoid unnecessary software subscriptions. Explain the opportunity cost of founder time and propose a zero-cash alternative when a paid activity is not informative enough.

RETURN
An itemized budget, spending gates and a lower-cost alternative plan.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find operational checks before offering a real pilot. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Business type: [type]. Location: [location]. Data collected: [types]. Delivery claims: [claims].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Create a bounded verification checklist for licensing, tax treatment, consumer terms, privacy, payment setup and product safety only where relevant to the described offer. Use current authoritative local sources if browsing is available; do not issue definitive legal or tax advice from assumptions. Distinguish requirements needed before taking payment from questions relevant only at scale. Assign unresolved items to qualified verification before dependent actions.

RETURN
A source-linked requirements checklist with priority, uncertainty and who should verify each item.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check what happens if your main channel or tool changes. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Critical platform: [platform]. Dependence: [details]. Revenue and cost exposure: [figures].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map dependence on traffic, accounts, pricing, integrations, payment processing and policy. Verify current official rules where needed. Create scenarios for lower reach, higher fees, limited access and outage without assigning invented probabilities. Identify customer relationships and assets I can retain lawfully outside the platform. Design a modest second-channel or delivery fallback test before spending on complex infrastructure.

RETURN
A platform-dependency risk table, three scenarios and a practical resilience experiment.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me identify what limits delivery before adding more customers. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Pilot records: [time, delays, errors]. Process: [steps]. Capacity target: [target].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Map throughput across steps and identify queues, rework and handoffs. Separate bottlenecks supported by data from plausible explanations. Estimate maximum capacity using observed rates while noting sample limits. Compare reducing scope, changing scheduling, improving one step and hiring, including cost effects. Design a small process test that measures end-to-end outcome rather than optimizing a step that does not constrain delivery.

RETURN
A bottleneck map, improvement options and an operational experiment with measurable acceptance criteria.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me turn test results into a proportionate decision. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Original hypothesis and criteria: [paste]. Results: [data]. Deviations: [changes].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Compare results with criteria set before the test. Audit eligibility, denominator, missing data, incentives, tracking and procedure deviations. Separate positive evidence from explanations that could also produce the result. State what was learned and what remains unknown. Do not rescue a weak result by redefining success after the fact. Recommend continue, revise, retest or stop with evidence and a next commitment sized to uncertainty.

RETURN
A test review, alternative-explanation table and a clearly bounded decision memo.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me find what a failed offer test does and does not show. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Visitors or contacts: [counts]. Eligibility: [details]. Funnel events: [counts]. Offer and tracking: [details].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Check whether eligible people saw, understood and could complete the offer before interpreting zero sales. Inspect tracking, payment, availability and technical problems. Separate weak traffic, weak message, low trust, poor timing and lack of value as competing explanations. Identify which are supported and which need another test. Avoid declaring either total failure or guaranteed potential from a tiny sample.

RETURN
A diagnostic tree in plain language, evidence gaps and the cheapest discriminating follow-up test.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me investigate whether encouraging results could be misleading. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Results: [data]. Incentives: [details]. Participant sources: [sources]. Founder involvement: [hours].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Audit purchases and usage for friend support, deep discounts, one-off novelty, subsidized delivery, incentives and heavy founder assistance. Distinguish a real benefit from scalable acquisition or delivery. Identify correlated observations from the same buyers. Propose a replication under more realistic conditions while preserving fair customer expectations. Explain what a stronger result would show and what would still remain unproven.

RETURN
A false-positive audit, adjusted conclusions and a replication plan.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me decide whether evidence supports another experiment. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Failed assumptions: [list]. Supporting evidence: [data]. Remaining budget: [amount]. Alternative direction: [idea].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Separate fixable execution problems from evidence that the customer problem, payment or delivery model is weak. Compare the cost and learning value of one focused revision with stopping or testing another concept. Include opportunity cost and sunk-cost bias. Require each proposed revision to state which evidence supports it and what could falsify it. Avoid endlessly changing the idea to escape a clear negative result.

RETURN
A revise/stop comparison, explicit decision criteria and a bounded next test only if justified.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me decide whether your evidence warrants product development. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Evidence portfolio: [notes and results]. Build cost: [estimate]. Remaining risks: [list].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Evaluate problem, buyer, payment, acquisition, delivery and repeat-use evidence separately. State which assumptions a build would test and which it would not. Compare a limited build against another manual pilot. Define the minimum build scope, budget cap and post-build acceptance criteria if proceeding. Do not present revenue, signups or positive interviews as a guarantee of product-market fit.

RETURN
A go/no-go memo, unresolved risks and a conditional minimum-build specification.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me check whether people return after the first interaction. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
User-level anonymized dates: [data]. Expected use cycle: [cycle]. Incentives and reminders: [details].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Define a meaningful repeat action and an appropriate observation window for the task. Group users by start date so recent users are not counted as failures prematurely. Separate organic returns from reminders and incentives where observable. Analyze cancellations, inactivity and changing need without inventing reasons. Explain small-sample limits and identify whether the offer needs recurring use to support its proposed business model.

RETURN
A cohort worksheet, repeat-use assessment and interview questions for returners and nonreturners.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me keep observations and decisions traceable. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Existing notes: [paste]. Hypotheses: [list]. Experiments: [list].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Create a ledger connecting each hypothesis to source, date, participant or dataset, observation, interpretation, limitations and next action. Keep raw facts separate from conclusions and do not overwrite earlier negative evidence. Add status fields such as untested, supported, challenged and unresolved with definitions. Track experiment deviations and decision changes. Show how an apparently strong conclusion might depend on one weak source.

RETURN
A populated example from my inputs, a blank ledger template and a maintenance routine.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me look for the strongest reasons your current interpretation could be wrong. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Current thesis: [thesis]. Evidence: [sources and data]. Decision under consideration: [decision].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Construct the strongest skeptical case across customer need, buyer authority, switching, price, delivery and acquisition. Tie criticisms to data or label them hypotheses. Do not use aggressive language as a substitute for analysis. Include the best evidence supporting the idea and explain which objection it answers. Rank unresolved objections by impact and propose a test capable of distinguishing the optimistic and skeptical explanations.

RETURN
A balanced red-team memo, evidence gaps and three decisive test options.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me increase commitment gradually based on what remains uncertain. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Pilot outcomes: [results]. Resources: [limits]. Remaining assumptions: [list].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Build a 30-day plan around the next riskiest assumptions rather than a full launch checklist. Include repeatable delivery, another customer source and retention where relevant. Define weekly outputs, spend caps and decision gates. Preserve manual steps until evidence justifies automation. Distinguish replication of the pilot from testing a new segment. Identify the finding that would pause the plan and the minimum evidence required for the next larger commitment.

RETURN
A weekly roadmap, measurement plan and explicit expansion gates.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.

Act as a business validation researcher. Help me explain what your research supports to a partner or yourself. Work within the constraints below and explain specialist terms in plain English.

MY INPUTS
Research timeline: [dates]. Tests and data: [records]. Audience for report: [audience].
Idea or offer: [describe it]. Target market and location: [who and where]. Available time and validation budget: [hours, deadline, currency and amount]. Existing evidence: [paste observations, anonymized notes, results or source links; write none if unavailable].

YOUR TASK
Summarize the original idea, assumptions, methods, sample, costs and results. Distinguish observed facts from interpretation and list counterevidence. Compare results with original criteria, disclose procedure changes and avoid extrapolating small samples to the market. State which risks remain and recommend a proportionate next step. Include source references and an appendix for raw measurements without confidential personal details.

RETURN
An executive summary, assumption-by-assumption evidence table, limitations, decision and next-test brief.

EVIDENCE RULES
Separate supplied facts, sourced findings, interpretations and untested assumptions. Never invent customers, quotes, market sizes, conversion benchmarks, competitor prices or test results. If browsing is available and useful, verify current claims and include source URLs and dates; otherwise state what needs checking. Ask up to five essential questions when missing inputs block the task; label provisional assumptions for everything else. Treat opinions and AI role-play as hypotheses, not customer evidence. End with the cheapest useful next action, the observation it should collect and what finding would change the decision. Propose decision thresholds before seeing results, explain their rationale and do not describe small convenience samples as representative.