Product-Market Fit

In brief

Marc Andreessen named the idea in a 2007 essay titled "The Only Thing That Matters," crediting the venture investor Andy Rachleff for the underlying framing. Product-market fit is being in a good market with a product that can satisfy that market. Andreessen's test was blunt and behavioral rather than statistical: you can always feel when it is not happening (customers are not getting value, word of mouth is not spreading, sales cycles drag, press reviews are lukewarm) and you can always feel when it is (usage grows as fast as you can add servers, money piles up, you are hiring sales and support as fast as you can). The claim underneath is that a startup's life divides into two phases, before fit and after fit, and that almost everything a founder should do depends on which phase they are in. Chasing growth before fit is the classic way to die.

The full treatment

The problem it answers

Startups fail far more often than they succeed, and founders wanted to know why. The intuitive answer, that they had a bad team or ran out of money, described the symptom rather than the cause. Rachleff and Andreessen argued that the dominant cause is building something the market does not strongly want, then spending as if it did. A great team with a mediocre market usually loses; a mediocre team with a great market often wins, because the market pulls the product out of them. Product-market fit gives founders a single question to answer before any other: is the market pulling, or am I pushing? Getting that question wrong is expensive, because every other decision (how much to raise, how fast to hire, how much to spend on marketing) is only correct on one side of the line.

How it works

Think of a threshold. Below it, the machine does not compound. You can double the sales team and revenue does not double, because the product does not retain the customers you win, so you are refilling a leaking bucket. Adding features does not help, because you are adding to something people did not want enough. Above the threshold the dynamics invert. Customers who try the product keep using it, tell others, and pull in more customers at low cost. Growth begins to feed itself. The founder's job flips from a search problem (what should we build, and for whom) to a scaling problem (how do we keep up, hire fast enough, and not fall over). Andreessen's practical rule follows directly: before fit, protect your runway and iterate cheaply; do not spend to scale a thing that is not yet working. After fit, spend aggressively, because the constraint is now your own capacity, not demand.

How you know you have it

The honest answer is that it is felt more than measured, but people have tried to make it concrete. Sean Ellis proposed a survey question: ask active users how they would feel if they could no longer use the product, and treat the product as having fit if at least 40 percent say they would be "very disappointed." The threshold is a rule of thumb, not a law, but it captures the right thing: fit lives in retention and in how much a real user would miss the product, not in signups or downloads, which are easy to buy. Other signals point the same way: cohorts of users who stay rather than churn, organic word of mouth, and demand that arrives faster than you solicited it. Vanity metrics (raw traffic, press mentions, total registered users) can all rise without any fit at all.

Distinctions that matter

Fit is a state, not an event, and it is not permanent. A market shifts, a competitor arrives, or the product outgrows the segment that first loved it, and fit can be lost. It is also market-specific: a product can fit one narrow segment and completely miss the next. Crucially, product-market fit is not the same as a good idea, a working product, or even happy early users who are friends of the founder. It requires a market with real, urgent demand and a product that genuinely serves it. This is why the concept is paired with the discipline of customer development from Steve Blank, and with jobs-to-be-done, which asks what job the customer is actually hiring the product to do. Fit is the destination; those are ways of searching for it without going bankrupt.

Lineage

The named concept is recent, but its roots run deeper. Andreessen credited Andy Rachleff, who developed the framing while at the venture firm Benchmark and later at Wealthfront, drawing in turn on the pattern that market quality dominates outcomes. The intellectual ancestor is Steve Blank, whose The Four Steps to the Epiphany (2005) reframed a startup not as a small version of a big company executing a known plan, but as an organization searching for a repeatable business model. Blank's customer development method (get out of the building, test whether anyone wants this, before scaling) is the process by which fit is found. Eric Ries, a student of Blank's ideas, turned this into The Lean Startup (2011) and its loop of build, measure, learn. Further back stands Joseph Schumpeter's picture of the entrepreneur as the disruptive agent who introduces new combinations the market did not know it wanted, an idea captured in creative destruction. Product-market fit is, in effect, the moment Schumpeter's new combination lands.

The strongest case for it

Its power is that it forces sequencing. Founders instinctively want to grow, and the market rewards visible momentum, so the temptation is always to spend on sales and marketing early. The concept says: do not, until the thing works, because scaling a product without fit converts money into a bigger leaking bucket. This diagnosis matches what investors see. Andreessen's stronger claim, that in a great market the product gets pulled out of the startup while in a bad market even a great product cannot save you, explains a genuine pattern: companies that looked mediocre have thrived on strong markets, and brilliant teams have died in weak ones. As a single organizing question for an early-stage founder, few frameworks are as clarifying. It also correctly locates the truth in behavior (do users come back and pull others in) rather than in the founder's hopes.

The strongest case against it

The main charge is that the term is vague to the point of being unfalsifiable. Andreessen himself said you feel it rather than measure it, which makes it easy to declare in hindsight and hard to act on in advance: any success can be relabeled as fit, any failure as its absence. Rahul Vohra and others have tried to rescue it with metrics like the 40 percent survey, but critics note that a single threshold is arbitrary and gameable. A second objection is that "fit" implies a binary switch, when in practice it is a matter of degree that can be partial, temporary, and segment-specific; treating it as a line crossed can lull a team into complacency the moment demand appears. Andy Rachleff has partly conceded this by emphasizing value hypothesis and growth hypothesis as separate things. Third, later practitioners (Brian Balfour prominent among them) argue the framework is incomplete because it omits the channel and the business model: a product can fit a market yet still fail if there is no viable, affordable way to reach that market or to make the unit economics work. On this view, product-channel fit and model-market fit matter as much as product-market fit, and the original slogan flatters founders into thinking the hard part is over once users are happy.

Where it stands now

The term is now standard vocabulary in startups and venture capital, taught in accelerators and repeated in board meetings. Its core insight, do not scale spending before demand is real, is close to consensus. At the same time, the sophisticated view has moved past the single slogan toward a family of fits (product-market, product-channel, channel-model) and toward measurable proxies for what was once purely intuitive. The honest current position is that product-market fit remains the most useful one-word diagnosis for whether an early company is being pulled or is pushing, while being an imperfect and easily abused one. Most experienced investors use it as Andreessen intended: not as a metric to optimize, but as a warning against the most common and most expensive startup mistake, which is spending to grow something the market does not yet want.

Test yourself

Picture a product you use and would genuinely miss. Now picture one you signed up for and forgot about within a week. The concept lives entirely in the gap between those two feelings. Ask of any early company you are tempted to admire: is the market pulling this out of them, or are they pushing it uphill with money and effort? If you cannot tell, notice that impressive signups, funding, and press coverage told you nothing about the answer.