Product market fit: what it is, how to measure it, and what it changes in GTM

Andreessen defined it in his 2007 essay "The only thing that matters", arguing that the market matters more than the team or the product. When a market badly wants what you sell, it pulls the product out of you: customers buy as fast as you can ship, usage grows, and you hire to keep up. When it does not, everything feels like pushing.

Two things get lost when people quote the definition. First, fit is always with a market, meaning a specific group of buyers you can name and count. A product can fit one segment and miss another. Second, it is not permanent. Competitors, pricing shifts and new technology can erode it, so the measures below are worth re-running every quarter.

The Sean Ellis test (the product market fit survey)

Sean Ellis, who ran early growth at Dropbox and LogMeIn, asks users one question: "How would you feel if you could no longer use this product?" The answers are very disappointed, somewhat disappointed, or not disappointed. After benchmarking nearly 100 startups, he found that companies where at least 40% of users answered "very disappointed" tended to grow, and those below it usually struggled.

The best-known worked example is Superhuman. Rahul Vohra wrote up for First Round Review how the email client scored 22% in 2017, rose to 33% by narrowing to its most enthusiastic segment, and reached 58% within three quarters by building what that segment asked for.

Aim for 40 or more responses before trusting the number, and remember it measures sentiment. It works best next to behavior data.

Retention curves

Plot the share of each signup cohort still active after one, two, three months and onward. Without PMF the curve keeps falling toward zero. With it, the curve bends and flattens: a stable group keeps using the product month after month. A flat curve at any level is a sign of fit with someone. How high it flattens tells you how big the business can get.

For reference, Lenny Rachitsky's survey of operators (2020) called about 60% six-month user retention good and about 80% great for SMB and mid-market SaaS, with enterprise products expected to hold more. For a sales-led product, track logo retention and net revenue retention by cohort instead of user activity.

Sales signals

In a sales-led company, PMF shows up in the pipeline before it shows up in surveys. Look at deals from the last two quarters and check whether the sales cycle is getting shorter, whether win rates are rising in one segment, and whether more deals arrive by referral or inbound without being pushed. If every deal needs a founder, a discount and a custom feature, you do not have it yet.

Most GTM mistakes at early companies are timing mistakes: scaling a motion before the product has earned it. A GTM strategy should look different on each side of PMF.

When to hire your first salesperson

Hire the first AE when founders have closed enough deals from the same segment that they can explain why those customers bought, what the objections are, and how long it takes. If the founders cannot write the playbook, a new hire will not find it for them. A common pattern is to hire two reps at once, so you can tell a bad hire from a bad playbook.

When outbound works

Outbound turns a known problem into meetings. It works once you have an ideal customer profile you can filter a database by and a problem buyers already know they have. A customer who looks like the prospect helps a lot. Before PMF it is still useful as research. Send small batches to a few segments, read every reply, and let reply rates by segment tell you where the pull is. Grade those tests on positive replies, since meetings are too few to compare.

Product-led growth (PLG) means the product itself does most of the acquiring: people sign up and get value on their own, then pay or invite teammates without talking to sales. Sales-led growth means a rep runs the evaluation, usually for larger deals with several people involved. PMF looks different in each. In PLG it shows up as activation and free-to-paid conversion. In sales-led, it shows up as win rates and retention by cohort.

The hybrid motion

Most B2B software companies that start product-led add sales once larger companies show up in the signups. Individuals adopt the product bottom-up, and sales closes the company-wide deal top-down. I wrote about why I think this is the default path for most B2B software in PLG is not a tactic. The hard part is knowing which accounts to call. That is what product usage signals are for: several active users at one company, a usage limit hit, or a visit to the pricing page, joined with firmographic fit and sent to a rep.

Sales-led companies can borrow from PLG as well: a free tool, a sample of the output, or a sandbox lets buyers see value before the first call, and the people who use it become your warmest outbound list.

Once you know which segment pulls, the next work is turning it into a system: a scored account list, plus the signals and outbound that decide who hears from you this week. The GTM playbook lays out that build step by step. At OneGTM I build those systems for B2B teams that have found their segment and need pipeline from it.