Product-market fit (PMF)

Also called product/market fit.

Business basics8min read

Product-market fit (PMF) is the match between a product and a specific market in which customers get enough value to keep using or buying it. It is evidence of demand, not a funding stage or a fixed numerical score.

Outside business, PMF can also mean probability mass function in statistics or primary myelofibrosis in healthcare.

Key takeaways

  • PMF belongs to a customer segment and use case, not necessarily your entire market.
  • Track retention by cohort. A growing customer total can hide the fact that most customers leave.
  • Sean Ellis's 40% "very disappointed" survey test is a useful benchmark, not a universal pass mark.
  • Raising money does not prove PMF. Investors can fund a promising search for it.
  • PMF does not guarantee profitability. Customers can love a product that costs too much to deliver.

How product-market fit works

A product has PMF when it satisfies a strong enough need for a defined group of customers. The evidence comes from what those customers do: buy, return, renew, expand their use or recommend it to others.

Start with three questions:

  1. Who gets the value?
  2. What job does the product do for them?
  3. What behavior shows that they want it again?

"Small businesses like our software" is too broad. "Independent accounting firms use our client-request tool throughout every monthly close and renew after their first year" is a claim you can test.

Early PMF is often narrow. One group retains well while another signs up enthusiastically and disappears. Averaging them together can lead you to build for the wrong customer.

The right evidence also depends on the product. Daily activity matters for a workplace messaging tool. It is a poor test for software used once during an annual compliance process. Measure return behavior at the frequency the underlying need occurs.

Founder-led sales do not disqualify a company from having PMF. But separate demand for the product from demand for unlimited access to the founder. If every customer needs a custom implementation and ongoing manual work, find out whether that work is part of the business you intend to build.

Measure behavior before declaring victory

No single metric proves PMF. Look for several signals that support the same explanation.

SignalWhat to measureWhat can mislead you
RetentionCustomers or users returning within comparable cohortsTotal active users rising because acquisition outpaces churn
Willingness to payPurchases, paid conversions and renewalsFree pilots or discounts that never expire
Depth of valueCompletion of the product's core taskLogins that produce no useful outcome
Customer pullReferrals, unsolicited requests and expansionPraise without continued use or spending

A cohort is a group that started during the same period. Comparing successive cohorts helps distinguish a better product from a larger marketing budget.

For subscriptions, account retention and net revenue retention answer different questions. Account retention measures how many customers remain. Net revenue retention, or NRR, measures how much recurring revenue remains from the starting customers after churn, contraction and expansion, excluding new customers.

You need both. A few expanding accounts can conceal widespread customer losses.

The PMF survey and the 40% benchmark

Sean Ellis popularized a survey test based on how customers would feel if they could no longer use a product. His benchmark is at least 40% answering "very disappointed."

The calculation is:

PMF survey score = "Very disappointed" responses ÷ all responses to that question × 100

If 22 of 50 respondents choose "very disappointed," the score is:

22 ÷ 50 × 100 = 44%

That is a signal to investigate, not a certificate to display.

Ask people who have had a reasonable chance to experience the product's core benefit. Report your eligibility rules, number invited, number responding and survey date. Do not remove indifferent answers to improve the result.

A survey of active customers tells you about people who stayed. It does not explain everyone who left. Pair it with retention data covering the full starting cohort.

In First Round Review, Rahul Vohra describes how Superhuman used this approach to identify its strongest customer segment and refine the product. The useful lesson is the segmentation and follow-up research, not just the headline percentage.

The same product retains 40% of consultants and 86% of agencies

Consider a fictional workflow software company charging $100 per account per month. It signs up 100 solo consultants and 100 small agencies in January.

Each segment starts with $10,000 in monthly recurring revenue. Over six months, some accounts cancel. Retained agencies also buy paid add-ons. There are no price changes, downgrades or reactivations, and the revenue calculations exclude new customers.

MetricSolo consultantsSmall agencies
Starting paying accounts100100
Starting monthly recurring revenue$10,000$10,000
Accounts remaining after one month8094
Accounts remaining after three months5588
Accounts remaining after six months4086
Six-month account retention40%86%
Six-month recurring revenue before expansion$4,000$8,600
Expansion revenue at month six$0$1,900
Six-month recurring revenue including expansion$4,000$10,500
Six-month NRR40%105%

For agencies, account retention is:

86 ÷ 100 × 100 = 86%

Their NRR is:

($10,000 − $1,400 lost to churn + $1,900 expansion) ÷ $10,000 × 100 = 105%

The company also ran its survey after the first month. Fifty active customers responded in each segment. Twelve consultants and 22 agencies answered "very disappointed," producing scores of 24% and 44%.

Both measures point toward agencies. They report stronger attachment, retain better and expand spending.

The founder's next move should be to investigate and serve that segment more deliberately, not declare that the whole small-business market has PMF. Interview retained and churned agencies, identify the valuable workflow, and test whether the next agency cohort behaves similarly.

The combined six-month account retention is 126 ÷ 200 = 63%. That blended figure hides the difference that matters most for product and sales decisions.

What the original essay says

In "The only thing that matters" (2007), Marc Andreessen describes PMF through customer demand: customers buying quickly, usage growing and the company struggling to keep up.

That description concerns market pull. It is not a requirement that every startup must suddenly experience explosive growth.

For a founder, the practical question is whether customers want the product strongly enough to support continued demand. A launch spike, a successful fundraising round or a large waitlist does not answer that question by itself.

In the example above, agencies provide stronger evidence of pull than consultants. Six months of results still leave open questions about annual renewals, acquisition costs and whether another cohort will perform as well.

What to put in a fundraising deck

Using the fictional example, an illustrative slide could read:

Small agencies are our strongest segment. Of 100 accounts acquired in January, 86 remained after six months. The cohort's NRR was 105%. In a month-one survey, 22 of 50 active agency respondents said they would be very disappointed without the product.

Below that, show the monthly retention curve, starting cohort size and survey method. Explain whether expansion came from more seats, add-ons or a price change.

A pre-seed company can present an early PMF hypothesis honestly. A company asking investors to finance faster acquisition needs stronger evidence that the customers it acquires will stay. The claim should match the maturity of the data.

What is normal, and what should you question?

There is no universal retention percentage that separates PMF from failure across software, marketplaces, consumer products and occasional-use services.

Ellis's 40% benchmark applies to his survey method. It is not a universal retention target, a fundraising requirement or a substitute for customer behavior.

Question a PMF claim when:

  • Only cumulative signups appear, with no retention cohorts.
  • Survey percentages omit the number of respondents.
  • Revenue growth depends on replacing customers who quickly leave.
  • A few large accounts conceal weak retention elsewhere.
  • Customers praise the product but will not pay or return.
  • Acquisition depends on discounts or manual service that the financial model ignores.

None of these automatically disproves PMF. Each identifies a question the evidence still needs to answer.

ConceptMain questionEvidence that helps answer it
Problem-solution fitDoes this solution address a real customer problem?Interviews, prototypes and successful early use
Product-market fitDoes a defined market want this product enough to sustain demand?Retention, renewals, repeat purchases and customer pull
GrowthIs the business getting bigger?Revenue, customers or usage increasing over time
ProfitabilityDoes revenue exceed expenses?Financial results for the relevant period

A company can grow without PMF by spending heavily to replace lost customers. It can also have strong customer demand without profitability because its delivery costs or acquisition spending are too high.

Questions people ask

What is the full meaning of PMF?

In business and startups, PMF means product-market fit: evidence that a product satisfies a strong need in a defined market. It describes customer demand, not a particular funding round.

What is PMF used for?

Founders use it to decide where to focus product development, which customers to pursue and whether to increase growth spending. Investors use PMF evidence to assess whether demand is durable enough to support a larger business.

Is there a PMF formula?

There is no universal formula for product-market fit. The familiar survey score divides "very disappointed" responses by all responses to that question. Retention and revenue formulas measure other parts of the picture; none alone establishes PMF.

Can a startup have PMF before revenue?

A free product can show strong demand through repeated use and retention. That supports a claim about user demand, but it does not prove that users will pay or that advertising, transactions or another revenue model will work.

Can a company lose product-market fit?

Yes. Customer needs, alternatives and purchasing behavior change. Watch whether newer cohorts retain worse, renewals become harder or customers stop using the core workflow. An old survey score cannot establish that the product still fits its market.

Related terms

Sources

  1. Marc Andreessen, "The only thing that matters," 2007.
  2. Sean Ellis, Startup Marketing, "The Startup Pyramid," 2010.
  3. Rahul Vohra, First Round Review, "How Superhuman Built an Engine to Find Product Market Fit," 2018.