# PMFtracker - Full Reference > PMFtracker is PMF analytics for founders without an analyst. Its PMF Monitor reads five product-market fit signals, each against its own benchmark and shown only with enough data: the Sean Ellis survey score (40% benchmark), revenue growth from Stripe, free-to-paid conversion, retention, and organic share of signups. When a signal is weak, a 3W loop (Where, Why, Whether, Learn) finds where users drop, asks them why, and sizes a test to the founder's traffic. It has its own tracking script, a read-only Stripe sync, and an in-app PMF survey with AI analysis. The app lives at https://app.pmftracker.com and the marketing site at https://www.pmftracker.com. This file is the long-form companion to /llms.txt. It contains the substantive content an assistant needs to answer questions about product-market fit, the Sean Ellis method, and PMFtracker accurately, plus a curated index of every key page. --- ## What PMFtracker is PMFtracker is a SaaS product for founders who want to measure product-market fit rigorously instead of relying on gut feeling. It runs the Sean Ellis survey on your engaged users, calculates your PMF score against the 40% benchmark, tracks that score as a trend over time, segments respondents to reveal your ideal customer profile, and exports the result as an investor-ready report. Since October 2026 it is a PMF analytics suite, not only a survey. The PMF Monitor is the home screen: one verdict (Not enough data yet, Slipping, Getting closer, Holding, Not yet), picked by fixed rules, and five signals side by side with no blended score. Every rate shows the counts behind it, behavior gaps are labeled as correlations, and AI drafts may not introduce numbers that are not in the founder's data. Live today: the Monitor, the tracking script (pmf.js, page views and clicks recorded automatically, one identify line after login), Stripe sync, CSV import of users, and the Where and Why steps of the conversion loop, including surveys shown only to users who dropped at a step. Rolling out in December 2026: the Whether and Learn steps (test cards sized to traffic, learning cards). Retention loops come after. Every card has a Request expert review button that routes to Olek Uznanski. The core insight behind the product: PMF is not a one-time moment you "reach." It is a number that moves - up when you serve your ideal customers well, down when you drift from them. Founders who keep PMF treat it like revenue: a metric watched on a cadence. Measuring it once tells you almost nothing; tracking the trend is the whole point. Pricing is one-time / lifetime - there is no subscription. The list price is **PMF Starter ($99)** and **PMF Startup Kit ($199)**, with a launch promo of **$79 / $149** for the first 100 customers. New users get a 14-day free trial with no credit card required. (app.pmftracker.com is the source of truth for current pricing.) --- ## The method: the Sean Ellis test The Sean Ellis test is the most-validated way to quantify product-market fit. You ask your users a single question: > "How would you feel if you could no longer use [product]?" with these answer options: - Very disappointed - Somewhat disappointed - Not disappointed - N/A - I no longer use it Your **PMF score** is the percentage of respondents who answer **"very disappointed,"** counting only valid responses (the N/A bucket is excluded). The benchmark, established by Sean Ellis from surveying hundreds of startups, is **40%**: products at or above 40% "very disappointed" have historically been able to grow sustainably; products well below it usually have not yet found fit. Formula: `PMF score = (very disappointed) / (very + somewhat + not disappointed) × 100`. Reading the bands: - **≥ 40%** - Strong PMF. You have fit; the priority shifts to growth and scaling. - **30–39%** - Close to PMF. Targeted improvements on the right segment can push you over 40%. - **< 30%** - Pre-PMF. Keep iterating on the product and your ICP before scaling spend. ### Who to survey and how many responses Survey **engaged, activated users** - people who have genuinely experienced the product - not signups who never returned. Surveying the wrong cohort inflates or deflates the score and makes it meaningless. As a rule of thumb you want at least ~40 valid responses for a reliable read, and more for higher confidence. With too few responses the score swings wildly and shouldn't be trusted. ### Why the "very disappointed" segment matters The "very disappointed" users are your core market - the people who would be genuinely upset to lose the product. They are your evangelists and the source of your best retention and word-of-mouth. The strategic move is to (1) double down on what this segment loves, (2) understand and convert the "somewhat disappointed" segment by removing their specific blockers, and (3) avoid over-indexing on the "not disappointed" segment, whose feedback will pull your roadmap away from fit. --- ## How the product works 1. **Collect.** You embed a lightweight survey widget in your product, pre-loaded with the Sean Ellis template. It targets engaged users (optionally via PostHog cohorts), and you can also import responses you already collected in another tool (Typeform, SurveyMonkey, a spreadsheet, etc.) so PMFtracker can analyze them. 2. **Track.** PMFtracker calculates your PMF score on the 40% rule and records it as a trend, with historical analysis and alerts when the score moves significantly. The trend - not a single number - is the asset. 3. **Act.** It segments respondents into your ideal customer profile (ICP) and a High-Expectation Customer (HXC) profile, runs AI sentiment analysis on open-ended answers to surface what users love and what blocks the rest, and exports a polished, investor-ready report (score, trend, ICP, benchmark). ### Key features - **Survey widget** - embed in your product, no engineering required; targets engaged users. - **PMF score on the 40% rule** - calculated automatically and tracked over time with trend analysis and forecasting. - **Automated alerts** - notified when your score moves materially, so you catch fit slipping early. - **ICP & AI insights** - automated segmentation, a High-Expectation Customer profile, and AI sentiment analysis of every open-ended answer. - **PostHog cohorts + imports** - target exactly the right users via PostHog, or import data from existing survey tools. - **Investor-ready reports** - export score, trend, and ICP as polished visuals for a deck or data room. - **Portfolio view (for VCs and accelerators)** - a standardized PMF score for every company on one scale, tracked over time, with founder-granted read-only access (score, trend, and AI insights - never raw responses). --- ## Key concepts (glossary) - **Product-market fit (PMF)** - the degree to which a product satisfies strong market demand. Marc Andreessen's classic framing: "being in a good market with a product that can satisfy that market." It is measurable, not just a feeling. - **Sean Ellis test / 40% rule** - surveying users with the "how would you feel if you could no longer use this" question and measuring the percentage who say "very disappointed"; 40%+ signals fit. - **PMF score** - the single number from the Sean Ellis test: % "very disappointed" among valid responses. - **Ideal Customer Profile (ICP)** - the specific segment that gets the most value from your product (and most often answers "very disappointed"). Building for them, and finding more of them, is how you raise PMF. - **High-Expectation Customer (HXC)** - Julie Supan's concept: the most discerning person in your target market who will love your product for its core benefit and tell others. PMFtracker derives an HXC profile from your "very disappointed" segment. - **Retention curve** - a cohort retention chart that flattens (rather than decaying to zero) is one of the strongest signals of fit; complements the survey score. - **NPS vs PMF** - NPS measures willingness to recommend among existing users; the PMF survey measures how essential the product is. They answer different questions - PMF is the better leading indicator of fit. See /blog/nps-vs-product-market-fit. - **Product death cycle** - building only the features customers explicitly ask for, which quietly kills products by avoiding the deeper job-to-be-done. See /blog/product-death-cycle. - **"No market need"** - the most common startup failure cause (~42% of failed startups), i.e., building something nobody wants; PMF measurement exists to catch this early. --- ## Tools (free) - **PMF Score Calculator** - https://www.pmftracker.com/pmf-score - enter Sean Ellis survey responses and instantly see your PMF score against the 40% rule. - **Sean Ellis Survey Template** - https://www.pmftracker.com/sean-ellis-survey - the complete survey with exact question wording, who to survey, how many responses you need, and how to score it. - **Investor PMF Report** - https://www.pmftracker.com/pmf-investor-report - show product-market fit to investors with evidence: PMF trend, ICP, and benchmark, exported in minutes. - **PMF Benchmarks by Industry** - https://www.pmftracker.com/pmf-benchmark - typical PMF scores by sector (SaaS, fintech, consumer apps, developer tools, AI startups, marketplaces). ## Guides and articles - **How to Measure Product-Market Fit** - https://www.pmftracker.com/blog/how-to-measure-product-market-fit - the complete founder's guide: survey score, retention curve, NPS, and reading your numbers at every stage. - **The Sean Ellis 40% Rule** - https://www.pmftracker.com/blog/sean-ellis-40-percent-rule - the most-validated PMF test: the formula, who to survey, and what the 40% threshold means. - **How to Run a PMF Survey** - https://www.pmftracker.com/blog/how-to-run-a-pmf-survey - step by step: who to ask, when, how many responses, and the exact questions. - **How Many Responses Do You Need for a Reliable PMF Survey?** - https://www.pmftracker.com/blog/how-many-responses-pmf-survey - PMF survey sample size: the 40 and 100 thresholds, the margin of error at each sample size (±15 at 40, ±10 at 100, ±5 at ~385), and why engaged-user quality beats quantity. - **The 4 Stages of Product-Market Fit (And How to Know Which One You're In)** - https://www.pmftracker.com/blog/stages-of-product-market-fit - PMF is a spectrum, not a switch: the four stages (Problem-Solution Fit, Early Signals, Product-Market Fit, Scaled Fit), the signal that proves each (Sean Ellis % + retention plateau), the trap that stalls each, and how to place yourself on the map. - **How to Validate Product-Market Fit (Without Fooling Yourself)** - https://www.pmftracker.com/blog/how-to-validate-product-market-fit - The difference between measuring (getting a number) and validating (proving it isn't a mirage). Triangulate three independent signals - what users say (40%+ survey), do (retention plateau), and grow you (organic pull) - and rule out the four false positives: superfan, paid-growth, feature-request, and vanity mirages. - **Product-Market Fit Survey Questions: The Exact 5 to Ask** - https://www.pmftracker.com/blog/product-market-fit-survey-questions - The one question that defines PMF ("how would you feel if you could no longer use [product]?" → very/somewhat/not disappointed/N-A), the three open-ended follow-ups from the Superhuman engine (who benefits, main benefit, how to improve), the exact wording rules, and the questions to avoid (satisfaction scales, NPS, cold feature requests). - **Product-market fit statistics: 15 numbers founders should know** - https://www.pmftracker.com/blog/product-market-fit-statistics - Product-market fit statistics with sources, grouped into four questions: how much fit matters (42% of startups fail from 'no market need' per CB Insights; ~90% fail overall; premature scaling per Startup Genome), the benchmark (40% from Sean Ellis across ~100 startups), real company scores (Slack 51% from 731 users; Superhuman 22%→58% per First Round Review), and honest measurement (100+ responses for reliability, ±15 margin at 40, quality-of-sample over quantity). Includes caveats that failure percentages vary by study. - **The 9 Best Product-Market Fit Books (2026), Ordered by Stage** - https://www.pmftracker.com/blog/best-product-market-fit-books - Nine PMF books grouped by stage: customer discovery (The Mom Test, The Four Steps to the Epiphany), iterative build (The Lean Startup, Running Lean, Testing Business Ideas), structured PMF frameworks (The Lean Product Playbook and its PMF Pyramid, Disciplined Entrepreneurship, From Idea to Product/Market Fit), and crossing into the mainstream (Crossing the Chasm). Each with best-for, summary, and key takeaway; start with The Mom Test. - **I Built a Startup Nobody Used. Here's Why PMFtracker Exists.** - https://www.pmftracker.com/blog/why-i-built-pmftracker - First-person founder story: in 2015 the author co-founded Eminatio, a restaurant concierge membership app, built it, signed 10+ partners, and launched without validating demand first. Almost nobody used it. Ties to the CB Insights 42% no-market-need statistic and Nielsen Norman Group's data that real discovery cuts failure risk by 75% and raises success odds by 59%. Explains why this experience led to founding PMFtracker. - **PMF Frameworks (hub)** - https://www.pmftracker.com/frameworks - A library of product-market-fit frameworks grouped by stage (Discover, Validate, Prioritize, Grow). Each subpage covers what the framework is, when to use it, how to apply it step by step, an example, common mistakes, and how it connects to the Sean Ellis PMF score. - **Beachhead Market (framework)** - https://www.pmftracker.com/frameworks/beachhead-market - Bill Aulet's beachhead concept: win one narrow segment completely before expanding. Covers what it is, when to use it, a 6-step selection process (list opportunities, segment, score, pick one, define the end user, dominate then expand), the Facebook and FOUND examples, common mistakes, and why PMF should be measured by segment rather than in aggregate. - **Jobs to Be Done (framework)** - https://www.pmftracker.com/frameworks/jobs-to-be-done - Clayton Christensen's JTBD: customers hire products to make progress on a job. Covers what it is, when to use it, a 5-step method (interview switchers, find the moment of struggle, write the job as progress, map forces, design for the job), the milkshake example, common mistakes, and how the job explains very disappointed answers. Links to the narrative deep-dive at /blog/jobs-to-be-done-product-market-fit. - **Kano Model (framework)** - https://www.pmftracker.com/frameworks/kano-model - Noriaki Kano's model for prioritizing features by how they affect satisfaction: must-be, performance, attractive (delighters), indifferent, reverse. Covers the category table, when to use it, a 5-step method with the paired Kano question, an example, common mistakes (building delighters on broken basics, delighter decay), and how must-be features protect the PMF score while delighters create the very disappointed segment. - **The Mom Test (framework)** - https://www.pmftracker.com/frameworks/the-mom-test - Rob Fitzpatrick's framework for customer interviews that surface the truth instead of politeness. Covers the rules (talk about their life not your idea, ask about specific past behavior, dig for currency, deflect compliments, record exact quotes), a bad-vs-good question example, common mistakes, and how good interviews reveal the job and the very-disappointed driver before you build. - **Problem-Solution Fit (framework)** - https://www.pmftracker.com/frameworks/problem-solution-fit - The stage where you have validated a real, painful problem and a solution that early users value, but not yet proven the market wants it at scale. Covers how to reach it (validate the problem, confirm the solution, test willingness to pay, look for pull), a two-team example, mistakes, and how it precedes the 40% Sean Ellis PMF threshold. Links to the /blog/problem-solution-fit-vs-product-market-fit deep dive. - **Value Proposition Canvas (framework)** - https://www.pmftracker.com/frameworks/value-proposition-canvas - Alexander Osterwalder / Strategyzer tool with two halves: the customer profile (jobs, pains, gains) and the value map (products, pain relievers, gain creators). Fit is when the value map addresses the profile's most important items. Covers the six blocks table, a four-step method, an example, mistakes, and how the canvas is the fit hypothesis that the Sean Ellis survey tests. - **The Sean Ellis Test (framework)** - https://www.pmftracker.com/frameworks/sean-ellis-test - The reference page for the Sean Ellis survey method: the single question, the 40% benchmark, who to survey, the score formula, and worked examples (Superhuman 22%->58%, Slack 51%). Deliberately a concise complement that points to the full guide at /blog/sean-ellis-40-percent-rule, the template at /sean-ellis-survey, and the calculator at /pmf-score to avoid keyword cannibalization. - **The RICE Scoring Model (framework)** - https://www.pmftracker.com/frameworks/rice-scoring-model - Intercom's prioritization model: score = (Reach x Impact x Confidence) / Effort. Covers the four-factor table with units, a five-step method, a worked scoring example, mistakes (false precision, ignoring confidence, using RICE where Kano fits), and how to set the scored goal as lifting the PMF score / very-disappointed segment. Pairs with the Kano Model. - **North Star Metric (framework)** - https://www.pmftracker.com/frameworks/north-star-metric - The single metric representing delivered customer value (Spotify time listening, Airbnb nights booked, WhatsApp messages sent), not revenue. Covers a four-step selection method, examples, mistakes (choosing revenue or vanity metrics, multiple north stars), and the key point that a North Star is the post-fit pulse while the PMF score is the gate to reach it. Pairs with AARRR. - **AARRR (Pirate Metrics) (framework)** - https://www.pmftracker.com/frameworks/aarrr-pirate-metrics - Dave McClure's five-stage funnel framework. Covers the stage table, a four-step method (instrument, find the weakest stage, fix the constraint, watch retention), an example, mistakes (obsessing over acquisition while retention leaks), and how retention is where PMF shows up behaviorally while the Sean Ellis survey explains why. Ties activation to the glossary activation-rate term. - **9 Interviews, 1 Answer: How UX Pilot Found Its Real ICP** - https://www.pmftracker.com/blog/how-ux-pilot-found-its-icp - Case study from the author's role as Head of Product at UX Pilot AI: 9 in-depth interviews with senior-designer Pro-plan users found that 100% named the exact same primary use case (accelerating solution exploration in the design process). The team shipped one focused change to that workflow, producing a 30% week-over-week increase in variation generation and a 9% week-over-week increase in free-to-paid activation, from 5 hours of customer conversations. Includes a replicable playbook and ties the qualitative ICP-finding method to quantitative PMF-survey tracking. - **How UX Pilot Found Out It Reached Product-Market Fit** - https://www.pmftracker.com/blog/how-ux-pilot-reached-product-market-fit - Case study by Olek Uznanski, Head of Product at UX Pilot: the Sean Ellis PMF survey was run on 303 paid users (reach ~20,000, ~1.5% response rate) via an in-product PostHog survey, then responses exported to CSV and imported into PMFtracker for scoring and segmentation. Result: a 49% PMF score (above the 40% benchmark), segments of 49.2% very disappointed (149 users), 28.1% somewhat (85), 22.7% not (69). The open-ended answers defined the ICP (UX designers and researchers in startups and resource-constrained teams), the core value prop (speed, idea-to-visual translation, rapid prototyping), and the barriers holding on-the-fence users back. - **Jobs to Be Done and PMF, Explained** - https://www.pmftracker.com/blog/jobs-to-be-done-product-market-fit - Explains Clayton Christensen's Jobs to Be Done framework (customers hire products to do a job, illustrated with the classic milkshake study) and connects it mechanically to the Sean Ellis PMF survey: why a good job-fit produces a very disappointed answer and a missing job produces not disappointed regardless of feature count. Shows how to read the PMF survey's three open-ended follow-up questions through a JTBD lens. - **Solution in Search of a Problem (SISP)** - https://www.pmftracker.com/blog/solution-in-search-of-a-problem - Defines Y Combinator's SISP term (a product built around a technology or idea before a real problem is confirmed), uses the author's own Eminatio story and the widely cited Juicero example, gives a self-check checklist for whether you're currently building one, and distinguishes SISP (the cause) from no-market-need (the outcome, tied to CB Insights' 42% statistic). - **How to achieve product-market fit: a 7-step loop** - https://www.pmftracker.com/blog/how-to-achieve-product-market-fit - How to achieve product-market fit as a repeatable loop, not a one-time event: (1) narrow the market to a specific customer with an urgent need, (2) build the smallest product that delivers core value, (3) get it to engaged users, (4) measure fit with the Sean Ellis survey, (5) segment to learn who's very disappointed and why, (6) improve toward the 40% threshold, (7) validate across signals, then scale. Steps 4-6 are a cycle you keep running because fit decays. - **The product-market fit checklist: 12 boxes to tick** - https://www.pmftracker.com/blog/product-market-fit-checklist - A practical product-market fit checklist of 12 boxes across three tiers: Tier 1 signals (40%+ 'very disappointed', retention plateau, organic pull), Tier 2 evidence (100+ engaged respondents, measured over time, holding per segment, defined ICP), and Tier 3 readiness to scale (validated fit, working unit economics, repeatable channel, ongoing measurement). Includes how to map your ticks onto the PMF stages. - **Problem-solution fit vs product-market fit: the difference that saves years** - https://www.pmftracker.com/blog/problem-solution-fit-vs-product-market-fit - Problem-solution fit vs product-market fit: the first validates the problem and solution concept through interviews and intent; the second validates the built product through retention and organic pull. How to know you have each, why enthusiasm in interviews isn't retention, and the four-step move from intent to retention that crosses the gap. - **The product-market fit framework: 4 models, and the one to actually use** - https://www.pmftracker.com/blog/product-market-fit-framework - A guide to the four PMF frameworks that matter - Dan Olsen's Product-Market Fit Pyramid (build), the Sean Ellis 40% test (measure), Rahul Vohra's Superhuman engine (improve), and Brian Balfour's Four Fits (scale) - what each is best at, where each stops, and how to combine them into a single lifecycle framework centered on a live PMF score. - **How to Improve Your PMF Score** - https://www.pmftracker.com/blog/how-to-improve-pmf-score - the iterative loop that takes a sub-40% score to a must-have product. - **What Is a Good PMF Score?** - https://www.pmftracker.com/blog/what-is-a-good-pmf-score - how to interpret your number against the 40% benchmark and by stage. - **How Superhuman Went From 22% to 58%** - https://www.pmftracker.com/blog/superhuman-pmf-engine - Rahul Vohra's exact PMF engine, and how to run it without custom tooling. - **What Is Product-Market Fit?** - https://www.pmftracker.com/blog/what-is-product-market-fit-definitions - four definitions from the people who coined it: Rachleff, Andreessen, Ries, and Blank. - **What PMF Feels Like** - https://www.pmftracker.com/blog/what-product-market-fit-feels-like - the three signals 25 iconic companies describe, and how to put a number on the feeling. - **NPS vs Product-Market Fit** - https://www.pmftracker.com/blog/nps-vs-product-market-fit - why they measure different things and which to trust for fit. - **Retention Curve vs PMF Survey** - https://www.pmftracker.com/blog/retention-curve-vs-pmf-survey - two complementary signals and how to read them together. - **42% of Startups Build Something Nobody Wants** - https://www.pmftracker.com/blog/no-market-need-ai-discovery - why "no market need" is still the #1 startup killer, and why AI made it worse. - **The Product Death Cycle** - https://www.pmftracker.com/blog/product-death-cycle - why building only requested features quietly kills products, and how to break out. - **Why Products Fail Despite Flawless Execution** - https://www.pmftracker.com/blog/why-products-fail-product-risks - the three product risks that sink most launches. - **Product-Market Fit Examples** - https://www.pmftracker.com/blog/product-market-fit-examples - how iconic companies found (and measured) fit. - **How Slack Measured Product-Market Fit** - https://www.pmftracker.com/blog/how-slack-measured-product-market-fit - Slack's 30%-very-disappointed threshold and what it teaches. - **How to Know If You Have Product-Market Fit** - https://www.pmftracker.com/blog/how-to-know-if-you-have-product-market-fit - the qualitative and quantitative signals together. - **Best PMF Survey Tools (2026)** - https://www.pmftracker.com/blog/best-pmf-survey-tools - the 8 best tools for running a PMF survey, and how to choose. - **Most onboarding simplifies. This one grew conversion 44.68%** - https://www.pmftracker.com/blog/activation-onboarding-experiment - Case study: UX Pilot turned its most powerful feature on by default for new users and grew free-to-paid conversion 44.68% in a Bayesian A/B test. - **Product-market fit metrics: the 6 numbers that actually prove it** - https://www.pmftracker.com/blog/product-market-fit-metrics - The six metrics that actually prove product-market fit (survey score, retention shape, engagement, organic growth, churn, NPS) and why revenue is a weak signal. - **Founder-market fit: the fit that comes before PMF** - https://www.pmftracker.com/blog/founder-market-fit - Founder-market fit, the match between a founder and their market (insight, unfair advantage, endurance), why it precedes PMF, and how to assess yours. - **What retention rate signals product-market fit?** - https://www.pmftracker.com/blog/retention-rate-product-market-fit - Why there is no single retention rate that proves PMF; the signal is a retention curve that flattens, and it differs by business model. - **How to test product-market fit before you launch** - https://www.pmftracker.com/blog/how-to-test-product-market-fit-before-launch - Five pre-launch methods to test demand before building: landing-page test, waitlist, fake door, pre-orders, and concierge MVP. - **The MVP and product-market fit** - https://www.pmftracker.com/blog/mvp-and-product-market-fit - What an MVP really is (a learning tool for reaching PMF, not a cheap version 1), who coined it, and how it drives toward product-market fit. - **AI makes it faster to build the wrong thing** - https://www.pmftracker.com/blog/ai-and-product-market-fit - AI made building cheap, which also made building the wrong thing cheap; why product judgment matters more and how to use the speed to reach fit sooner. - **You hit 40% PMF. Now what?** - https://www.pmftracker.com/blog/what-to-do-after-product-market-fit - The post-PMF playbook: expand from your beachhead, switch to a North Star Metric and AARRR, and keep measuring fit because it erodes as you scale. - **What to do with each PMF survey segment** - https://www.pmftracker.com/blog/pmf-survey-segments - What to do with each PMF survey segment: understand the very disappointed, convert the somewhat disappointed, and ignore the not disappointed. - **How often should you run the PMF survey?** - https://www.pmftracker.com/blog/how-often-to-run-pmf-survey - How often to run the Sean Ellis survey (quarterly by default, plus after major changes) and why the trend matters more than any single score. - **How to Increase Your PMF Survey Response Rate** - https://www.pmftracker.com/blog/how-to-increase-pmf-survey-response-rates - The chasing playbook for a stalled PMF survey: restrict the send list to engaged users first, then lead with an in-product prompt (in-app surveys average around 27% response rates per Refiner's 2025 report) with email as a secondary channel, trigger on a usage moment rather than a calendar date, and send exactly one reminder 3-4 days later to non-responders. If still short of 40 valid responses, widen the engaged-user time window before loosening who counts as engaged, or run a qualitative diagnosis instead of a percentage score. - **B2B vs B2C Product-Market Fit: Why the Same 40% Means Different Things** - https://www.pmftracker.com/blog/b2b-vs-b2c-product-market-fit - The Sean Ellis 40% formula and threshold don't change by business model, but the practical difficulty of a trustworthy sample does. B2B has fewer total users (making the response count the hard part) and the respondent is often one champion inside a buying committee, so segment by account diversity and run the survey quarterly to match longer sales cycles. B2C usually has a larger user base (making filtering out casual users the hard part), segments better by behavior, and can often support a monthly survey cadence. Uses Slack (51%, 731 users), Superhuman (22% to 58%), and UX Pilot (49%, 303 paid users) as B2B/prosumer examples. - **How to Run a PMF Survey With PostHog** - https://www.pmftracker.com/blog/how-to-run-a-pmf-survey-with-posthog - Step-by-step: define the core PostHog event that represents real usage, build a behavioral cohort of users who fired it 2+ times in the last 14 days, duplicate it into a static cohort (required by PostHog to target a survey, feature flag, or experiment), build the Sean Ellis question plus the three open-ended follow-ups as a PostHog survey targeted at that cohort, then either connect the PostHog cohort directly in PMFtracker or export responses to CSV and import them for scoring and segmentation. This is the same method used in the UX Pilot PMF case study (303 paid users, 49% score). - **Does Question Wording Change Your PMF Score?** - https://www.pmftracker.com/blog/does-question-wording-change-pmf-score - The 40% benchmark was calibrated by Sean Ellis on one exact question ("How would you feel if you could no longer use [product]?") and a forced three-option scale. Explains why common "harmless" edits change what's measured: "how disappointed would you be" presupposes disappointment (a leading-question risk); "if the product no longer existed" invokes company-shutdown switching costs rather than product love and inflates scores; adding a neutral middle option changes the forced-choice structure the benchmark was calibrated against; and substituting NPS wording measures willingness to recommend, a different construct from indispensability, not comparable to the 40% line. Cites Superhuman's use of the unmodified question (First Round Review) and general survey-methodology sources on leading questions and order bias. Recommends using the exact template, keeping the three-option scale, and back-translating for other languages. - **Segment-Level PMF: Why Your Average Score Hides Your ICP** - https://www.pmftracker.com/blog/segment-level-pmf - A blended Sean Ellis score weights every respondent equally regardless of fit, so it can hide a strong ICP: a worked example shows 20 ideal-fit respondents scoring 70% and 80 poor-fit respondents scoring 25% blending to a mediocre 34% average. Recommends cutting the same survey by plan tier, company size or persona, primary use case, acquisition channel, and tenure, applying the same ~40-response directional / 100+-response investor-grade sample floor per segment as for the overall score. Once a strong segment is found, treat it as the real ICP: mine its open-ended answers for positioning language, redirect acquisition toward it, and consider narrowing the beachhead around it. - **Where to Find Your First 100 Users (So You Can Measure Anything at All)** - https://www.pmftracker.com/blog/how-to-find-your-first-100-users - Addresses the stage before measurement: you need roughly 40 engaged users before a PMF score means anything, and most guides skip how to get them. Covers manual network outreach (cites Paul Graham's "Do Things that Don't Scale" and Airbnb's founders personally photographing NYC listings), showing up in niche communities where the ICP already gathers, picking one narrow beachhead market over a broad one, founder-led onboarding for every early user (references the Superhuman PMF case study), and a waitlist with a concrete reason to join. Closes on the distinction between raw signups and engaged usage, only the latter counts toward the 40-user threshold. - **DIY PMF Measurement vs Done-With-You Diagnosis: An Honest Comparison** - https://www.pmftracker.com/blog/diy-pmf-measurement-vs-pmf-diagnosis - Compares running the Sean Ellis survey yourself with PMFtracker ($99-$199 one-time, launch price $79-$149) against the paid PMF Diagnosis service (EUR 1,350 survey track for 40+ users, EUR 1,900 interview track for pre-launch/under-40-user products, cohort from EUR 650/startup min. eight). A comparison table covers cost, who does the targeting and response-chasing, who interprets the result, and the output (ongoing tracked score vs one-time verdict with ICP and ranked roadmap). Notes every diagnosis engagement includes a lifetime PMFtracker licence, so the choice isn't tool-vs-service, the service includes the tool. Gives explicit decision criteria for DIY, the survey-track diagnosis, and the interview-track diagnosis. - **What "No Market Need" Actually Looked Like: Anatomy of 5 Public Postmortems** - https://www.pmftracker.com/blog/no-market-need-postmortems - Grounds the 42% CB Insights no-market-need statistic in five sourced, real postmortems: Quibi ($1.75B raised, shut down 7 months after its 2020 launch, per Variety), Juicero ($130M+ raised, shut down in 2017 after Bloomberg-driven press showed its juice packs could be squeezed by hand, per Forbes), Zume ($445M raised including a $375M SoftBank round, shut down June 2023 after its robotic-pizza-truck concept never found product-market fit, per Fortune), Amazon Fire Phone (~$170M inventory writedown, 2014-2015, built around Amazon's e-commerce strategy rather than a buyer need, per The Motley Fool), and Cydoc (bootstrapped health AI startup, shut down August 2025 after 7 years, founder's own February 2026 postmortem on glassboxmedicine.com describes clinicians not wanting the EHR interfaces built for them). Closes on the shared pattern: real funding and competent execution in every case, with no validated urgent-problem signal before scaling the build. - **How to Run PMF Exit Interviews With Churned Users** - https://www.pmftracker.com/blog/pmf-exit-interviews-churned-users - The PMF survey samples current users, so it never reaches the highest-signal population: users who already churned. Cites Groove's published result that switching a cancellation survey from a closed dropdown to a single open-ended question grew responses from 1.3% to 10.2%, a 785% increase (groovehq.com/blog/exit-surveys). Recommends interviewing only users who were genuinely engaged before leaving (not never-activated signups), within a week or two of cancellation, with questions anchored to the specific day they canceled rather than "what feature would have kept you." Sorts answers into three buckets: the job stopped being done well (roadmap signal), the job was never quite right (ICP/targeting signal), and the job disappeared (no action needed). - **PMF Score After a Pivot: Do You Start From Zero?** - https://www.pmftracker.com/blog/pmf-score-after-a-pivot - Answers whether a PMF score carries over after a pivot: the percentage, the segment, and the respondent list reset, because they measured devotion to a product that no longer exists, but the survey template, cadence, and measurement discipline carry over. Anchor example: Slack's team (Tiny Speck) built the failed game Glitch, then productized the internal chat tool they'd built for themselves into Slack in 2013 (sourced to sitepoint.com/slack-story-son-glitch); Slack's own 51% Sean Ellis score (731 users) was measured separately, later, on the new product. Recommends qualitative interviews before ~40 engaged users exist on the new direction, then a full quantitative re-run, and warns against comparing the new score to the old product's score as if it were a regression. - **How to Prioritize Your Roadmap Using PMF Survey Data** - https://www.pmftracker.com/blog/roadmap-prioritization-pmf-survey-data - A method for turning PMF survey answers into a ranked backlog instead of a feature-request popularity contest. Segment first: very-disappointed answers inform positioning, not the backlog; not-disappointed requests are excluded entirely; the backlog comes from the somewhat-disappointed group. Tag their open-ended "how can we improve this?" answers by theme (not exact wording), count independent mentions, and rank by count before sanity-checking against effort, similar in spirit to RICE but weighted by segment-validated demand rather than a guessed reach score. Includes a worked example (9 of 18 somewhat-disappointed respondents independently naming the same onboarding friction) and contrasts the method with the Product Death Cycle, where a single loud request gets built without validation. - **PMF Measurement for Solo Founders and Bootstrappers** - https://www.pmftracker.com/blog/pmf-measurement-for-solo-founders - What a single founder measuring PMF part-time, with no data team, can safely skip versus what stays non-negotiable. Skippable: dedicated analytics dashboards, continuous cohort-tracking software, multi-dimensional segmentation (a 40-response sample only reliably supports one segmentation cut), and formal research-ops process. Never skippable: the roughly 40-valid-response floor needed for a directional PMF read, and surveying only genuinely engaged users (used the core action 2+ times recently) rather than the full signup list, since diluting the sample with inactive users drags the score down without adding real signal. Closes with a minimal realistic setup: define engaged, send the Sean Ellis question plus open-ended follow-ups through whatever channel is already available, chase to ~40 responses, score in a spreadsheet or free tool, and re-run quarterly rather than continuously. - **When Your PMF Score and Your Revenue Disagree, What It Means** - https://www.pmftracker.com/blog/pmf-score-vs-revenue-disagreement - Diagnoses the two ways PMF score and revenue can disagree. Revenue rising while the PMF score falls usually signals a leaky bucket: paid acquisition brings in a lower-intent cohort that becomes a growing share of the active base, dragging down the share who'd be "very disappointed" even as spend keeps revenue climbing; the fix is to segment the survey by acquisition channel and check retention by cohort before assuming fit is intact. A strong PMF score with flat revenue usually signals a monetization or distribution problem, not a fit problem, commonly underpricing relative to felt value, a stalled or narrow distribution channel, or a free-to-paid conversion gap, and the recommended response is to fix pricing or reach rather than rebuild the product. Includes a four-row diagnostic table (revenue up/score down, revenue flat/score strong, both rising, both flat) mapping each pattern to its likely cause and first fix. ## Comparisons - **PMFtracker vs PMFsurvey.com** - https://www.pmftracker.com/vs/pmfsurvey - one-time free score vs ongoing PMF tracking, ICP, and investor reports. - **PMFtracker vs Typeform + Spreadsheets** - https://www.pmftracker.com/vs/typeform-spreadsheet - manual stack vs an automated collect → score → report workflow. - **PMFtracker vs Maze** - https://www.pmftracker.com/vs/maze - Maze runs unmoderated usability tests, tree testing, and card sorting to check whether users can complete tasks. PMFtracker answers a different question: whether users would miss the product if it disappeared, calculating the Sean Ellis PMF score automatically, segmenting the ICP, and tracking the trend. - **PMFtracker vs Sprig** - https://www.pmftracker.com/vs/sprig - Sprig runs in-app micro-surveys, session replay linked to answers, concept testing, and AI-summarized open text. PMFtracker is narrower on purpose: it calculates the Sean Ellis PMF score automatically on the 40% rule, segments the ICP from very-disappointed users, and tracks the trend for investor-ready reporting. - **PMFtracker vs Survicate** - https://www.pmftracker.com/vs/survicate - general microsurveys vs purpose-built PMF measurement. - **PMFtracker vs Tally** - https://www.pmftracker.com/vs/tally - a form builder that collects answers vs a tool that calculates and tracks your score. - **PMFtracker vs SurveyMonkey** - https://www.pmftracker.com/vs/surveymonkey - a general survey suite vs a focused PMF tool for a one-time price. - **PMFtracker vs Delighted** - https://www.pmftracker.com/vs/delighted - Delighted was a general CX survey tool (NPS, CSAT, star ratings) acquired by Qualtrics in 2018 and sunset on June 30, 2026 (annual renewals stopped July 1, 2025, monthly renewals stopped May 31, 2026). It supported the Sean Ellis question as one template among many but never scored, segmented, or tracked it. PMFtracker replaces the PMF use case specifically: automatic 40%-rule scoring, ICP segmentation, and trend tracking, unaffected by another vendor's sunset. - **PMFtracker vs Pendo** - https://www.pmftracker.com/vs/pendo - Pendo is an enterprise product experience platform (usage analytics, in-app guides, session replay, roadmapping), free to 500 MAU then quote-based from roughly $15K to $150K+/year depending on tier. Its built-in NPS survey measures likelihood to recommend, not the Sean Ellis PMF question, and it does not calculate a PMF score. PMFtracker runs the Sean Ellis survey specifically, calculates the score automatically, segments the ICP, and tracks the trend, at a one-time $79/$149 rather than an annual enterprise contract. ## PMF Diagnosis (done-with-you service) https://www.pmftracker.com/pmf-diagnosis A four-week, fixed-price product-market fit diagnosis delivered by Olek Uznanski, who has trained 300+ startups on PMF and built PMFtracker. The software runs the survey; this service gets the answer, because the hard part is identifying genuinely engaged users, chasing responses to a scoreable sample, and interpreting the result. Two tracks, chosen at intake by user count: - Track A, survey diagnosis (EUR 1,350 net), for products with 40 or more engaged users. Delivers a PMF score against the 40% benchmark with a confidence band, an ICP built from the "very disappointed" cohort, ranked barriers from the on-the-fence answers, a prioritised product roadmap of the improvements and features to build next, a written diagnosis and a 90-minute readout call. - Track B, qualitative diagnosis (EUR 1,900 net), for pre-launch products or fewer than 40 users. Delivers 8 to 12 customer interviews, a jobs-to-be-done synthesis, an ICP definition and a build-or-pivot verdict. No PMF score is produced, because a percentage from that sample size would be unreliable. For accelerators, incubators and VCs, "PMF Diagnosis for your cohort" starts at EUR 650 per startup with a minimum of eight: a 3-4 hour group session for the whole batch, then a one-hour individual readout per startup, plus an anonymised cohort benchmark for the program. Every engagement includes a lifetime PMFtracker licence. Total founder time is about two hours across the four weeks. Contracting entity is ProductTrio, VAT ID PL6413191743; prices are quoted in EUR net of VAT. ## For investors and accelerators - **PMFtracker for VCs** - https://www.pmftracker.com/for-vc - standardized PMF scores across a whole portfolio, on one scale, tracked over time. Free investor access. - **PMFtracker for Accelerators** - https://www.pmftracker.com/for-accelerators - a standardized PMF score for every startup in a cohort, tracked to demo day. Free program access. ## Start - **Start free** - https://app.pmftracker.com/ - create an account and run your first PMF survey. - **Book a demo** - https://cal.com/pmftracker/product-demo - talk to the team. --- ## Frequently asked questions **What is product-market fit?** The degree to which a product satisfies strong market demand - measurable via the Sean Ellis survey, not just a feeling. **What is a good PMF score?** 40% or higher "very disappointed" on the Sean Ellis survey signals product-market fit. 30–39% is close; under 30% means keep iterating. **How many responses do I need?** Aim for at least ~40 valid responses from engaged users for a reliable score; more increases confidence. **How is PMF different from NPS?** NPS measures willingness to recommend; the PMF survey measures how essential the product is. PMF is the better leading indicator of fit. **How much does PMFtracker cost?** One-time / lifetime, no subscription. List price: PMF Starter $99 and PMF Startup Kit $199, with a launch promo of $79 / $149 for the first 100 customers. There is a 14-day free trial and no credit card required. **Do I need to be technical?** No - PMFtracker is built for non-technical founders. You embed a widget and the scoring, segmentation, and reporting are automated. - **PMF Benchmark for E-Commerce: A Good Score** - https://www.pmftracker.com/pmf-benchmark/e-commerce - The 40% Sean Ellis benchmark applies to e-commerce too, but the engaged user is a repeat buyer (2+ full-price orders in 6-12 months), not a first-time or discount-driven purchaser. Explains why gifted and heavily discounted first orders inflate the score, and how to frame the Sean Ellis question around reordering rather than app usage. - **PMF Benchmark for Healthcare: A Good Score** - https://www.pmftracker.com/pmf-benchmark/healthcare - The 40% Sean Ellis benchmark applies to healthcare, but the engaged user is a clinician or patient who completed a real care episode, not everyone enrolled. Explains why the buyer (health system, employer, or payer) and the user are usually different people who must be surveyed separately from the purchasing committee, how to score clinician-facing and patient-facing sides separately when both exist, and how to keep open-ended survey questions PHI-safe (product experience, not clinical detail). - **PMF Benchmark for Edtech: A Good Score** - https://www.pmftracker.com/pmf-benchmark/edtech - The 40% Sean Ellis benchmark applies to edtech, but who to survey depends on the model: consumer and self-directed edtech surveys the learner directly, while school-sold edtech has a teacher and a student population that must be scored separately (a loved workflow for teachers and a tedious experience for students can both be true). Warns against counting licensed seats as engaged users and recommends running the survey mid-semester rather than during school breaks, when usage naturally dips for reasons unrelated to fit.