# Customer Discovery Prover MCP MCP > Customer Discovery Prover audits your startup pitch against real-world market rigor. It forces you to prove every claim—from who your customers are to how much they'll pay—using specific interview evidence, not just assumptions. Stop building for ideas; start building for proven pain points. ## Overview - **Category:** productivity - **Price:** Free - **Tags:** customer-discovery, product-market-fit, mom-test, icp, persona, willingness-to-pay, startup, customer-development ## Description Building a product based on gut feeling is the fastest way to fail. This MCP changes that process entirely by acting as a brutally honest reviewer of your market research. Instead of accepting generalized claims like 'the market needs' or defining a customer group simply as 'SMBs,' this tool forces you to ground every piece of data in real conversation. It checks if your persona is based on specific names, roles, and observed behaviors—not just demographics. You also learn the difference between someone saying they *might* pay for something (verbal interest) and showing genuine commitment like a signed Letter of Intent or scheduling a paid pilot. Because it requires you to submit evidence tied to past behavior using The Mom Test methodology, it quickly uncovers weak assumptions that would otherwise kill your startup before launch. When integrated into Vinkius, this MCP becomes the first checkpoint for any major product hypothesis. ## Tools ### validate_customer_discovery This tool audits a pitch's customer discovery claims, verifying if the persona, problem evidence, validation method, segments, and willingness-to-pay are all rigorously proven. ## Prompt Examples **Prompt:** ``` Our target customer is tech-savvy millennials who value productivity. The market clearly needs better project management tools. We asked 20 people if they would pay for this and 85% said yes. ``` **Response:** ``` PERSONA_INVENTED — 'Tech-savvy millennials who value productivity' is a demographic, not a persona. Who specifically? What role? What company? What is their specific pain — not 'productivity,' a behavior you observed? Also: 'Would you pay for this?' is a leading question (Mom Test violation) and '85% said yes' is biased validation, not WTP. This discovery fails on 4 of 5 axes. ``` **Prompt:** ``` Persona: 23 interviews with engineering managers at B2B SaaS companies (50-200 employees). Named: Sarah K. (VP Eng, fintech, 180 people), Mike R. (Eng Manager, healthtech, 85 people), Lisa T. (CTO, edtech, 60 people). Problem: 19/23 spend 6+ hours/week reconciling sprint data across Jira, Slack, and spreadsheets. Sarah: 'I lose every Monday morning to copying numbers between tools. It's 2026 and I'm doing manual data entry.' Current workaround: $2,400/year on Zapier + custom scripts, still takes 4 hours/week. Methodology: past behavior only — 'Walk me through your last sprint planning. What took the longest? What did you try to fix it?' Falsification: 'What would make this tool NOT worth switching for?' 8/23 said switching cost is the barrier. Segments: Segment A — VP Eng at 100-200 person companies, buying trigger is board reporting pressure, budget $500-1000/month, 6-week sales cycle. Segment B — Eng Managers at 50-100 person companies, buying trigger is team growth pain, budget $200-400/month, 2-week self-serve. WTP: 3 signed LOIs (Sarah K., Mike R., one unnamed), 2 paid pilots at $400/month starting June 15, Sarah agreed to be a launch case study. ``` **Response:** ``` DISCOVERY_PROVEN — Rigorous customer discovery validated. Persona: 23 named interviews, 3 specific people cited with roles and company context. Problem: 19/23 confirmation with quoted pain and $2,400/year workaround cost. Methodology: past behavior questions, falsification attempted (8/23 cited switching cost). Segments: VP Eng (100-200, board pressure, $500-1000) vs. Eng Manager (50-100, growth pain, $200-400). WTP: 3 LOIs, 2 paid pilots at $400/month with start dates, case study commitment. This discovery survives investor scrutiny. ``` **Prompt:** ``` Persona: 15 interviews with operations managers at manufacturing companies. Problem: 12/15 report spending 10+ hours/week on inventory reconciliation. Methodology: past behavior — 'Describe your last inventory audit. What went wrong?' Segments: We target SMBs in manufacturing. WTP: Strong interest from 8 companies, 3 requested demos. ``` **Response:** ``` SEGMENTS_CONFLATED — Persona, problem, and methodology pass. But segments fail: 'SMBs in manufacturing' is one label, not segmented. A 20-person machine shop and a 180-person food processor have different inventory systems, compliance requirements, budgets, and decision makers. Separate by: company size, sub-industry, regulatory requirements, and buying process. Also: WTP fails — 'strong interest' and 'requested demos' are not commitment. Show deposits, LOIs, or pilot agreements with dates. ``` ## Capabilities ### Validate customer personas It confirms if your target user profile is built on specific interview data and observable behaviors rather than general demographics. ### Evidence stated problems You prove that the pain point exists by citing specific quotes, costs, and current workarounds mentioned in customer interviews. ### Apply unbiased validation methods It forces your pitch to ask about past behaviors instead of leading questions about future intentions. ### Separate buyer segments You define distinct, actionable groups of customers that have unique budgets and buying processes, moving beyond generic labels like 'SMBs'. ### Test genuine willingness-to-pay It demands measurable commitment signals—like deposits or pilot dates—instead of relying on simple positive feedback. ## Use Cases ### Pitching a B2B SaaS solution A founder submits their pitch based on 'general enterprise pain.' The agent flags that they have conflated segments and demands the founder separate the needs of a 50-person agency from those of a 500-person manufacturer, each with different budgets. ### Validating an internal product idea A PM wants to launch a new feature for their existing client base. They run the concept through the MCP and discover that the assumed pain point is actually solved by a free, cheap competitor tool, requiring a fundamental pivot. ### Reviewing academic market research A researcher uploads interview data intended for an article. The MCP flags weak WTP signals, advising them that 'strong interest' isn't enough evidence and they need to ask for specific commitment metrics. ## Benefits - Stops you from building for demographics. You'll learn to define a persona by citing specific names, roles, and observed behaviors, not just 'busy professionals.' - Forces evidence-based problem statements. Instead of claiming 'the market needs X,' you must provide quotes detailing who said what, how often it happens, and the current cost of the workaround. - Eliminates biased questioning. It trains you to use The Mom Test by focusing only on past behavior—what users *did*—instead of asking about future promises. - Prevents lumping buyers into one group. You define distinct buyer segments based on different budgets, buying triggers, and decision-making processes. - Requires commitment signals. It elevates your pitch from 'they said they would pay' (verbal interest) to verifiable proof like LOIs or scheduled paid pilot dates. ## How It Works The bottom line is you get an objective audit of your market research, showing exactly what you need to prove before building anything. 1. Submit your current discovery claims, including assumed personas, problem statements, and proposed pricing models. 2. The MCP runs these claims through a rigorous five-point audit, cross-referencing every assumption against established best practices (Mom Test, evidence grounding). 3. You receive a detailed verdict report that flags exactly which assumptions are weak, where your data is biased, or which segments you’ve lumped together. ## Frequently Asked Questions **Does it conduct interviews?** No. It validates that your discovery process is grounded in real data — interview evidence, unbiased methodology, separated segments, and commitment-based WTP signals. It does not replace conversations with customers. It forces you to prove you had them. **What is The Mom Test?** A framework by Rob Fitzpatrick for conducting customer interviews that produce truthful data. The core rule: never ask leading questions about the future ('Would you use this?'). Instead, ask about past behavior ('When did you last encounter X? What did you do?'). People lie about future behavior — past behavior is a reliable signal. **Can pre-product startups use this?** Yes — it is designed for pre-product discovery. WTP signals for pre-product include: signed LOIs, paid design partnerships, deposits against future delivery, time commitments (agreed to a weekly feedback session), and reputation commitments (willing to be named as a design partner). You do not need a product to test willingness-to-pay. **How does using validate_customer_discovery handle sensitive interview data?** It processes your inputs in context only; Vinkius doesn't store private conversation details long-term. The tool evaluates the rigor of your discovery claims against best practices, but it never retains the specific names or quotes you provide. **What format must I use when running validate_customer_discovery?** You need a structured prompt that includes defined components: named personas, observed pain points with frequency, and concrete evidence of commitment. Simply stating 'the market needs' won't pass the validation. **If I try to make conflicting claims in validate_customer_discovery, what happens?** The tool's consistency engine catches contradictions immediately. For instance, if you claim unbiased validation but use leading questions, it rejects your input and provides Mom Test coaching. **Can I use validate_customer_discovery for industries other than B2B SaaS?** No; the methodology is universally applicable. The tool focuses on rigorous discovery principles—like separating buyer segments or testing WTP—which apply whether you're in agriculture, manufacturing, or tech. **What are the performance limits for calling validate_customer_discovery?** Usage is governed by your Vinkius subscription plan. The tool processes each validation request sequentially, allowing deep analysis before returning a definitive 'DISCOVERY_PROVEN' or failure verdict.