What is product-market fit? A guide for founders

TL;DR:

Product-market fit occurs when a product satisfies strong demand within a particular market, evidenced by active user engagement and word-of-mouth growth. Founders should evaluate PMF through behavioral metrics and the Sean Ellis survey, aiming for at least 40% “very disappointed” responses among engaged users. Clear communication and a focused visual identity are essential in early stages to accurately capture and accelerate product-market fit.

Product-market fit (PMF) is the state in which a product satisfies a strong, demonstrable demand within a specific market — and you can feel it, as Marc Andreessen famously described, when customers are buying and using your product as fast as you can possibly support them. Before you reach that point, almost everything else is noise. After it, almost everything accelerates.

If you are a founder or product manager asking whether you have PMF right now, check these signals first:

A substantial share of your engaged users say they would be “very disappointed” if your product disappeared, according to Sean Ellis’s survey heuristic
Week-4 retention cohorts are flat or rising, not declining toward zero
A meaningful share of new users arrives through word-of-mouth, with no paid acquisition
Churn is low and stable, not something you are constantly firefighting
Customer support volume is rising because usage is growing, not because the product is confusing

The single most important next step: run the Sean Ellis survey on users who have completed your core workflow at least twice. Their answers will tell you more than any analytics dashboard.

What is product-market fit, and where did the idea come from?
Why PMF changes everything about how you should run your startup
How do you measure product-market fit concretely?
Common mistakes that stop founders finding PMF
A repeatable framework for testing and improving PMF
What PMF looks like in practice: illustrative examples
How to spot when you are losing PMF
A one-page checklist to evaluate your PMF status right now
Product-market fit vs product and market validation
Key takeaways
The part most founders overlook about PMF
How clear visual identity shortens your PMF loop
Useful sources and further reading

What is product-market fit, and where did the idea come from?

The concept has a clear lineage. Andy Rachleff, co-founder of Benchmark Capital and later Wealthfront, coined the term in the early 2000s while teaching at Stanford. His central argument was that market selection precedes product success: a great team in a poor market will still fail, while a mediocre team in a great market can succeed. Rachleff framed PMF as the outcome of validating a value hypothesis — a precise answer to who the customer is and why they need this specific solution — before ever scaling with a growth hypothesis.

Marc Andreessen brought the phrase to a wider audience in 2007 through his blog series on startups. His framing is more visceral than Rachleff’s: PMF is the moment the market pulls the product out of the startup, not the moment the startup pushes the product into the market. That distinction matters because it shifts the founder’s job from persuasion to response.

Dan Olsen, author of The Lean Product Playbook, added the most operationally useful structure: the Product-Market Fit Pyramid. His framework layers five elements in a specific sequence:

Target customer — who, precisely, are you building for?
Underserved needs — which of their needs are currently unmet or poorly met?
Value proposition — how does your product address those needs better than alternatives?
Feature set — which capabilities deliver that proposition?
UX — how does the experience make the value self-evident?

The pyramid’s logic is that you must nail the base layers before investing in features or polish. Founders who skip straight to UX refinement without validating the target customer and underserved need are building on sand. Wikipedia’s summary reinforces this: PMF is a spectrum, not a binary switch you flip once and forget.

Why PMF changes everything about how you should run your startup

PMF is the fork in the road that determines whether you should be optimising for learning or for scaling. Before PMF, spending on growth channels, large engineering teams, or brand campaigns is almost always premature. After PMF, the cost of not scaling is the real risk.

Dimension	Before PMF	After PMF
Resource allocation	Minimal burn; focus on experiments	Invest aggressively in growth
Hiring focus	Generalists who can iterate fast	Specialists who can scale functions
Growth approach	Narrow, manual, qualitative	Broad, paid, and channel-diversified
Primary metrics	Retention, activation, qualitative feedback	Revenue growth, CAC, LTV, NPS

Pro Tip: Do not let a spike in sign-ups convince you that you have PMF. Check whether those users are returning and completing the core workflow. Behavioural metrics and unit economics must both support growth before you scale.

How do you measure product-market fit concretely?

The Sean Ellis 40% survey

The most widely used PMF metric is a single survey question: “How would you feel if you could no longer use this product?” with response options ranging from “very disappointed” to “not disappointed.” If 40% or more of respondents choose “very disappointed,” that is a strong signal of PMF. Below 40%, you are likely still searching.

Critically, the survey only produces reliable data when you send it to the right people. Sample only users who have experienced the core value of your product — those who have completed the primary workflow at least twice. Surveying all sign-ups, including people who never activated, will dilute your score and mislead your decisions. Aim for a sufficient number of valid responses before drawing conclusions.

Behavioural metrics to track alongside the survey

Stated intent and actual behaviour diverge constantly. Experienced product managers combine the Ellis survey with hard behavioural data:

Week-4 retention cohort: Plot the percentage of users still active four weeks after sign-up. A curve that flattens above zero indicates a retained core; a curve that trends toward zero suggests the product has not yet found its audience.
Month-3 retention: For subscription or recurring-use products, month-3 retention is a stronger signal of genuine habit formation.
Churn rate: For established SaaS products, monthly churn in the 5–7% range is a common reference point; anything significantly above that warrants investigation before scaling.
Organic referral growth: Track what percentage of new users arrive without paid acquisition. Rising organic share is one of the clearest PMF signals.
Usage frequency and task completion: Are users returning on their own schedule, or only when prompted? Are they completing the core workflow, or dropping off mid-task?

Metric	Target signal	Red flag
Sean Ellis score	≥40% “very disappointed”	—
Week-4 retention	Curve flattens above zero	Curve trends to zero
Monthly churn (SaaS)	5–7% or below	—
Organic referral share	Growing month-on-month	Flat or declining
Core workflow completion	40% or more of engaged users	—

Signal triangulation with qualitative follow-ups

After the Ellis survey, ask one open follow-up to “very disappointed” respondents: “What is the main benefit you get from this product?” Their answers reveal your real value proposition, often in language more precise than anything your team has written. Group those answers by theme. The dominant theme is your PMF signal; the outliers are noise. Prioritise the feedback from your most engaged users — the ones who would be most disappointed — and treat requests from casual users with caution.

Common mistakes that stop founders finding PMF

Most PMF failures are not product failures. They are targeting, measurement, or timing failures. Watch for these:

Targeting an overly broad market — “Anyone who shops online” is not a target segment. PMF appears first in a hyper-narrow group with an urgent, specific need.

The red flags that indicate you are mistaking traction for PMF: growth that spikes after a press mention and then collapses, high acquisition costs that require constant top-up, and low repeat usage after the first session. Each of these points to the same corrective action — go narrower, not broader.

A repeatable framework for testing and improving PMF

Y Combinator’s view is that PMF is discovered through rapid iteration, not perfected in a first release. The loop below gives you a repeatable structure:

Narrow your market. Choose the smallest segment with the most urgent version of the problem. Resist the pull toward a broad addressable market at this stage.
State your value hypothesis. Write one sentence: “We help [specific customer] achieve [specific outcome] better than [current alternative] because [specific reason].” This is your experiment anchor.
Build the smallest testable version. An MVP that delivers the core value proposition — nothing more. Features that do not directly test the hypothesis are distractions.
Recruit engaged users. Find people who match your target customer profile and have the problem you are solving. Do not rely on friends or passive sign-ups.
Measure with the Ellis survey and behavioural cohorts. Run the 40% survey after users have experienced core value. Track week-4 retention. Combine both signals before drawing conclusions.
Analyse qualitative follow-ups. Read every “very disappointed” response. Group themes. Identify the value proposition your fans actually describe.
Decide: iterate or pivot. If your score is below 40% but rising and the qualitative themes are converging, iterate on the same hypothesis. If the score is flat or falling and themes are scattered, pivot the value hypothesis or the target segment.

Applying Dan Olsen’s PMF Pyramid in practice

Olsen’s pyramid gives each step of the loop a structural home. In early experiments, spend 70% of your effort on the bottom two layers — target customer and underserved needs — before touching features or UX. A common mistake is inverting this: founders polish the interface before they have confirmed the customer and the need, which produces a beautiful product nobody wants.

Decision criteria for moving to scale: your Ellis score is at or above 40%, week-4 retention is stable, organic referral share is growing, and unit economics show a positive or near-positive payback period. All four, not just one or two.

Pro Tip: Write your value hypothesis on a card and pin it where your team can see it. Every feature decision, every design choice, and every experiment should be tested against it. If a proposed change does not directly strengthen the hypothesis, defer it.

What PMF looks like in practice: illustrative examples

Classic PMF stories share a recognisable pattern: a narrow initial segment, a product that solved one problem better than anything else available, and growth that arrived without heavy paid acquisition.

Early social platforms found PMF within a single university campus or city before expanding. The signal was not total user count but the density of daily active use within that narrow group. The lesson: depth of engagement in a small segment beats breadth of sign-ups across a large one.
Productivity tools that achieved PMF typically saw their core users adopt the product into daily workflows within the first week and refer colleagues without prompting. Organic referral was the leading indicator, not marketing spend.
A UK-relevant sketch for fashion and beauty founders: imagine a direct-to-consumer skincare brand targeting women aged 28–40 in London with sensitive, reactive skin. The brand’s PMF signal was not its total subscriber count but the fact that customer retention in beauty and wellness was driven almost entirely by repeat purchase and word-of-mouth recommendation within that specific demographic. The brand’s value proposition — formulations designed specifically for reactive skin, with full ingredient transparency — was the theme that dominated every qualitative follow-up. Expanding to a broader “all skin types” positioning before locking in that core segment would have diluted the signal entirely.

First Round’s research on PMF stages describes a progression from nascent to developing to strong to extreme fit. Most founders are operating somewhere in the nascent or developing range and mistake early traction for strong fit. Knowing which stage you are in sets realistic expectations and prevents premature scaling.

How to spot when you are losing PMF

PMF is not a permanent state. Stripe’s guidance is explicit: market conditions and customer expectations shift, and teams that stop measuring will miss the drift until it becomes a crisis.

Warning signs to watch:

Retention cohorts that were once flat begin declining across successive months
Support ticket volume rises faster than usage growth, suggesting the product is becoming harder to use or less relevant
Organic referral rates fall while paid acquisition costs rise
Activation funnels flatten: users sign up but fewer reach the core workflow
Your Ellis score drops below 40% in a re-survey of the same engaged segment

Recovery plan: re-narrow your segment to the users who are still most engaged. Re-run the value hypothesis experiment with that group. Focus product effort on core workflow completion and retention improvements before touching anything else. Do not add features during a PMF recovery phase — they add complexity without addressing the root cause.

Pro Tip: Keep a live PMF dashboard with your Ellis score, week-4 retention, organic referral share, and churn rate updated monthly. A single glance should tell you whether fit is holding or drifting.

A one-page checklist to evaluate your PMF status right now

Use this checklist before your next product or growth decision:

Have you surveyed only users who completed the core workflow at least twice?
Is your Ellis score at or above 40%?
Is your week-4 retention curve flattening above zero?
Is month-3 retention stable or improving?
Is a meaningful share of new users arriving through organic referral?
Is monthly churn within an acceptable range for your category?
Have you read and grouped the qualitative follow-ups from “very disappointed” respondents?
Do your unit economics show a viable payback period at current acquisition costs?
Have you re-surveyed within the last 90 days to check for drift?

Metric	Target threshold	Action if below threshold
Sean Ellis score	≥40% “very disappointed”	Narrow segment; re-run value hypothesis
Week-4 retention	Curve stable above zero	Investigate activation drop-off
Monthly churn (SaaS)	5–7% or below	Identify top churn reasons qualitatively
Organic referral share	Growing month-on-month	Survey fans; strengthen referral mechanics

Survey your most engaged users, not your most recent sign-ups. Recency bias in sampling is one of the most common reasons founders get a falsely low or falsely high Ellis score.

Product-market fit vs product and market validation

These three concepts are related but distinct, and conflating them is a common source of founder confusion.

Market validation is the earliest stage: you are confirming that a problem exists, that people experience it, and that they would pay for a solution. It requires no working product — interviews, landing page tests, and pre-orders all count. A validated market tells you the opportunity is real; it does not tell you that your specific product captures it.

Product validation confirms that your solution works as intended: users can complete the core workflow, the technology is reliable, and the value proposition is intelligible. A validated product is functional; it is not necessarily one that people will choose over alternatives or return to habitually.

Product-market fit is the synthesis of both: a specific product, matched to a specific segment, delivering a specific value that users find irreplaceable. The Ellis survey, retention cohorts, and organic growth are the evidence. You can have a validated market and a validated product and still lack PMF — which is why the measurement loop described above is not optional. It is the only way to know.

Key takeaways

Product-market fit requires a specific product matched to a specific segment, confirmed by behavioural retention data and the Sean Ellis 40% survey, before any scaling decision is justified.

Point	Details
Measure on engaged users only	Survey users who completed the core workflow at least twice; all sign-ups will dilute your score.
The 40% threshold is your anchor	An Ellis score below 40% means iterate; at or above 40%, combined with stable retention, you can consider scaling.
PMF is a maintained state, not a milestone	Re-survey every 90 days and keep a live dashboard; market drift can erode fit silently.
Narrow before you broaden	PMF appears first in a hyper-narrow segment; expanding too early destroys the signal.
Milda accelerates PMF discovery	Clear visual identity and UX make your value proposition self-evident to early users, shortening the feedback loop.

The part most founders overlook about PMF

The conventional wisdom says PMF is primarily a product problem. Build the right features, get the right retention numbers, and you have arrived. That framing is incomplete, and for founders in fashion, beauty, and lifestyle categories, it can be actively misleading.

PMF is partly a communication problem. A product can deliver genuine value and still fail to achieve fit because early users cannot immediately understand what it does, who it is for, or why it is different. When a user lands on your product for the first time, they are making a split-second decision about whether to invest attention. If your visual identity, your UX, and your messaging do not instantly communicate the core value proposition, you lose that user before the product ever gets a chance to prove itself.

This is particularly acute in consumer categories. A skincare brand with a genuinely superior formulation for reactive skin will still struggle to find PMF if its visual identity signals “mass market” rather than “specialist.” The product’s value is real; the communication of that value is broken. The Ellis survey will return a low score not because the product fails but because the wrong users are reaching it and the right users are not staying long enough to experience the core value.

The practical implication: during PMF discovery, strip your visual assets down to the essentials and focus every design decision on making the value proposition unmistakable. This is not the moment for brand storytelling or aesthetic complexity. It is the moment for clarity. Once you have confirmed fit in a narrow segment, you can layer in the richer brand world. Before that, every visual element should answer one question: does this help the right user understand, in under five seconds, that this product is for them?

How clear visual identity shortens your PMF loop

Founders running PMF experiments often underestimate how much friction a weak visual identity introduces into the feedback loop. When early users cannot immediately read what a product does or who it serves, they disengage before experiencing the core value, and your Ellis survey captures their confusion rather than the product’s genuine potential. The result is a falsely low score that sends you iterating on the wrong variables.

Milda works with fashion, beauty, and lifestyle founders at exactly this stage. The studio’s process combines visual strategy, identity design, UX direction, and full-stack website execution into a single, coordinated project — so your value proposition is communicated clearly from the first user touchpoint. A focused brand identity and a well-structured UX reduce the cognitive load on early users, which means more of them reach the core workflow, more of them experience the value, and your PMF data becomes cleaner and faster to act on.

If you are preparing for a PMF experiment or re-launching after a pivot, explore how UX direction supports early-stage startups and consider what a focused visual identity could do for your conversion and retention rates. Get in touch with Milda to discuss a scoped identity or UX project built around your current experiment stage.

Useful sources and further reading

The sources below are the most credible starting points for going deeper on product-market fit. Each one is worth bookmarking.

Marc Andreessen’s original PMF essay (pmarchive.com) — the canonical definition and the “before PMF / after PMF” framing that every subsequent writer has built on. Read this first.
12 things about product-market fit (a16z) — Andy Rachleff’s thinking on value hypothesis vs growth hypothesis, and why market selection precedes product success. Essential for founders at the earliest stage.
The real product-market fit (Y Combinator) — YC’s practical take on iterative MVP experimentation and letting the market pull the product. Useful for founders who are still searching.
PMFtracker: 7-step loop to achieve PMF — the most practical implementation guide for the Sean Ellis survey, including sampling guidance and follow-up question design. Download the survey template from here.
What is product-market fit (Stripe) — Stripe’s guide is particularly strong on treating PMF as an ongoing loop rather than a one-time milestone. Read the section on monitoring for drift.
Levels of PMF (First Round) — First Round’s staged model (nascent, developing, strong, extreme) helps founders calibrate where they are and set realistic expectations for what comes next.
Finding product-market fit in the tech industry (Harvard Business School Online) — a well-structured academic perspective that complements the practitioner sources above.
Product-market fit (Wikipedia) — a useful reference for the 40% heuristic and the spectrum framing of PMF, with links to primary sources.
Client retention in beauty businesses (Xolapp) — practical retention and loyalty tactics for consumer brands in beauty, directly relevant to PMF measurement in that category.

Table of Contents