Product-market fit

What is product-market fit?

Product-market fit is a milestone, not a plan. Marc Andreessen popularized the phrase in a 2007 essay, describing it as being in a good market with a product that can satisfy that market. The useful part is the felt experience: before fit you are pushing, and after fit the market pulls.

Pushing looks like this: you chase every lead personally, demos go well and then go quiet, people sign up and never return, and each customer costs as much effort as the last. Pulling looks different. Usage grows in weeks you ran no campaign, support tickets are about limits rather than confusion, customers introduce you to peers unprompted, and your problem shifts from finding demand to serving it.

Two clarifications. Fit is per segment: you can have it with solo consultants and none with agencies that look similar on paper. And it is not permanent. Markets shift, competitors improve, and a product that fit two years ago can stop fitting while revenue still looks fine.

How to tell whether you have it

The strongest evidence is a retention curve that flattens. Plot what share of each signup cohort is still active after one, three, and six months. If the line keeps sliding toward zero, you do not have fit however good this quarter's signups look. If it declines and then levels off, some group has genuinely adopted the product, and that group is your market.

Supporting signals: a rising share of signups arriving through word of mouth, shorter sales conversations, customers upgrading unprompted, and users complaining loudly when the product breaks. Indifference is the real enemy. A survey heuristic attributed to Sean Ellis asks active users how they would feel if they could no longer use the product, with roughly 40 percent answering "very disappointed" often cited as a sign of fit. Treat it as a rough gauge, and only run it once you have enough active users to survey.

Why product-market fit matters for startups

It decides what you should be doing this month. Before fit, the only work that counts is talking to users and changing the product. Marketing spend and sales hires are premature, because scaling a product people leave just loses money faster.

After fit, the priorities invert. Distribution becomes the constraint, and the risk is failing to serve demand you already created. Getting this order wrong is among the most expensive mistakes a small team can make, and it is common because hiring a marketer feels more actionable than admitting the product is not there yet.

Fit, strategy, and expansion

Concept	What it is	Question it answers
Product-market fit	The milestone you reach	Does a real market want this?
Product-market strategy	The plan for getting there	Which market, what positioning, what price?
Product-market expansion	What you do afterward	Which segment or product comes next?

Keeping these separate prevents a common error: treating a detailed product-market strategy document as evidence of fit, or starting product-market expansion into a second segment before the first one holds.

Product-market fit in practice

Say you built a document tool aimed at everyone. After a year you have 4,000 signups, $3,000 in monthly revenue, and a cohort curve that keeps falling. Sort accounts by retention, though, and one group stands out: roughly 60 independent bookkeepers who log in weekly and almost never cancel.

You interview a dozen of them, cut three features they never touch, add the two things they ask for repeatedly, and rewrite the homepage to speak only to bookkeepers. Signups drop, since the message now excludes most visitors. Revenue climbs anyway, churn falls, and half of new accounts arrive on a colleague's recommendation. Narrowing did not shrink the business, it revealed where the fit already was.

Common mistakes

• Declaring fit from a good month. One strong month of signups is not fit. Cohort retention rate over several months is.
• Scaling spend too early. Paid acquisition applied to a leaky product turns a slow problem into a fast one.
• Averaging away the signal. Blended numbers hide the one segment that loves you. Segment before you conclude anything.
• Believing interviews over behavior. Enthusiastic calls are not customer validation. Payment and repeat usage are.
• Pivoting too often. Constant direction changes mean no version ever gets a real test. Decide in advance what result would justify a pivot.

The path there is unglamorous: ship a minimum viable product, watch what a narrow group does with it, and keep cutting until one segment cannot do without you.