Signs You Have Product-Market Fit (And Signs You Don't)

The Signs That Matter

1. Flat Retention Curves

Your cohort retention curves flatten at 20–40% after 90 days. Users don't trickle away — a core group stays permanently. This is the strongest single signal of PMF.

The shape of the curve matters more than the number. There are two patterns:

The decay curve (no PMF): 100% → 60% → 30% → 12% → 5% → approaching zero. Every month, more users leave than stay. The product is a novelty. Even if you're adding 1,000 new users per month, the bucket is leaking faster than you're filling it.

The flat curve (PMF): 100% → 60% → 35% → 28% → 26% → 25%. The curve flattens. A core group never leaves. They've integrated your product into their workflow. The exact percentage doesn't need to be high — a flat curve at 25% means you've found product-market fit for 25% of your users. Now you need to find what makes that 25% different from the 75%.

The benchmark: B2B SaaS companies with strong PMF typically show 90-day account retention of 25–40% for their ICP segment. Companies without PMF show 90-day retention below 15% — and still decaying.

How to measure it: Plot signup cohorts by week. For each cohort, calculate what percentage of users are still active at day 7, day 30, day 60, and day 90. A cohort is "active" if they've completed a meaningful action — not just logged in, but used the core feature. If the curve flattens for any cohort, you have PMF for that cohort's segment.

A note on sample size: You need 30–50 active users per cohort to see a statistically significant pattern. Fewer than 30 and the curve is noise — one power user can skew the entire result.

2. Organic Referrals

15%+ of new signups come from word-of-mouth. Customers are actively recommending your product to peers without being asked. You can trace referrals back to specific customers.

This is the hardest false positive to generate. You can't buy organic referrals. You can't fake them with a referral program (those are incentivized, not organic). When someone recommends your product to a peer, they're putting their own reputation on the line. That's a costly signal — and that's why it's reliable.

The benchmark: Companies with strong PMF see 15–30% of new signups coming from unattributed direct traffic, word-of-mouth, or "a colleague recommended it." If your referral rate is below 10%, either your product isn't remarkable enough to share, or you're acquiring through channels that don't produce referral-prone users.

How to measure it: Add a "How did you hear about us?" field at signup. Track the percentage who select "a colleague" or "word of mouth." Cross-reference with UTM data — direct traffic with no UTM source is often organic referral. If you can't trace it, you're probably undercounting.

3. Pricing Power

When you raise prices by 10–20%, fewer than 5% of customers churn because of the price increase. The product's value exceeds the price increase. Customers complain but stay.

Pricing power is the ultimate market validation. If you can raise prices without losing customers, the market is telling you that your product delivers more value than you're capturing. If you raise prices and 15% of your base leaves, your product is a commodity — priced at the market rate, not the value rate.

The benchmark: SaaS companies with PMF see fewer than 5% incremental churn from a 10–20% price increase. The customers who leave are typically at the margin — they were already considering leaving, and the price increase was the final trigger. Companies without PMF see 12–20% churn from the same increase.

The test: Raise prices for new customers only. If existing customers don't notice the change for 6–12 months (because they're grandfathered), you haven't actually tested pricing power. The real test is raising prices for existing customers with 90 days' notice — and watching who stays.

4. Expansion Revenue

NRR above 110%. Existing customers pay more over time through plan upgrades, seat additions, or usage increases. Your revenue grows even without new logo acquisition.

Net Revenue Retention (NRR) tells you whether your existing customer base is growing or shrinking. NRR above 100% means expansion revenue exceeds churn and contraction. NRR above 110% means your product is becoming more valuable to customers over time — the definition of PMF.

The benchmark by segment:

• SMB (under $10K ACV): 105–115% NRR is strong. SMB customers expand slowly and churn faster.
• Mid-market ($10K–100K ACV): 110–125% NRR is strong. Mid-market customers expand through seat additions and feature upgrades.
• Enterprise ($100K+ ACV): 120–140% NRR is strong. Enterprise customers expand through multi-year contracts and platform adoption.

How to measure it: NRR = (Starting MRR + Expansion MRR − Contraction MRR − Churned MRR) / Starting MRR. Calculate it monthly. Track it by customer segment, not just in aggregate. If your overall NRR is 105% but your enterprise segment is at 130% and SMB is at 92%, you have PMF for enterprise — and a problem with SMB.

5. Shortening Sales Cycles

For ICP-fit prospects, sales cycles are getting shorter over time. What took 3 months at $1M ARR takes 6 weeks at $3M ARR. The market is pulling the product in, not being pushed.

Shortening sales cycles mean your product's reputation is doing the selling for you. Prospects come in already convinced — they've heard from peers, read case studies, or evaluated competitors and concluded your product is the right fit. The sales process becomes confirmation, not persuasion.

The benchmark: Companies with PMF see sales cycle length decrease by 20–40% over 12–18 months as their product-market fit strengthens. If your sales cycles are staying flat or getting longer as you scale, your product isn't pulling prospects in — your sales team is pushing harder to compensate.

The caveat: This only applies to ICP-fit prospects. If you're broadening your ICP as you scale (which most companies do), your average sales cycle will naturally lengthen. Segment your data by ICP fit, not by all deals.

The Signs That Lie

1. High NPS

NPS measures sentiment, not retention. A customer can NPS 9 and still churn at renewal. NPS correlates with PMF only when combined with retention data. Alone, it's a feel-good metric.

Why it lies: NPS asks "how likely are you to recommend us?" — a hypothetical question about future behavior. People are terrible at predicting their own future behavior. They genuinely intend to recommend your product, and then they forget, get busy, or find something better. A study of 7,000 consumers found that NPS predicts actual referral behavior with only 0.3 correlation — meaning 91% of the variance in actual referrals is unexplained by NPS alone.

When NPS is useful: When combined with retention data. If your NPS is 50+ AND your retention curve is flat, NPS confirms what retention already told you. If your NPS is 50+ but your retention curve decays, NPS is a lagging indicator of customer satisfaction — not a leading indicator of customer behavior.

2. Press Coverage

Journalists write about novelty, not necessity. Being featured in TechCrunch doesn't mean customers need your product. It means your product is interesting. Interesting ≠ essential.

Why it lies: Press coverage measures your newsworthiness, not your usefulness. A product can be genuinely innovative (worthy of press) and still not solve a problem that customers will pay to solve repeatedly. The graveyard of well-funded, well-covered products with flat retention curves is long.

The test: Track what happens to your signup rate in the 30 days after a press feature. If the spike decays back to baseline within two weeks, the press was noise. If the baseline permanently lifts, the press introduced your product to a segment that actually needed it.

3. Investor Enthusiasm

Investors fund potential. PMF requires evidence. A funded company with flat retention has runway, not PMF.

Why it lies: Investors are incentivized to believe in your potential — they've already allocated capital and reputation to your success. Their enthusiasm reflects their own commitment, not your product's market fit. A seed round from a top-tier fund is a vote of confidence in the team and the market, not evidence that the product fits the market.

The pattern we see repeatedly: Companies with $10M+ in seed funding and 90-day retention below 15% tell themselves they have PMF because "our investors believe in us." The investors believe in the team's ability to find PMF — not that PMF already exists.

4. A Long Feature Request List

Feature requests don't indicate PMF. They indicate engagement — but engagement with the idea of your product, not necessarily the product itself. Customers who love a product don't request 50 features. They request 2 and use everything else.

Why it lies: A long feature request list signals that customers are thinking about your product — but it doesn't tell you whether they're using it. In fact, a high volume of feature requests often correlates with low adoption: customers who aren't getting value from the core product imagine that additional features will solve their problem. Customers who are getting value request specific, targeted improvements.

The signal in the noise: Look at the type of feature requests, not the volume. "Can you add X?" is neutral. "We're using X every day but it breaks when Y happens" is a strong signal — it means they're using the core feature deeply enough to hit edge cases.

5. Total User Count

"One million users!" means nothing without activation data. If 90% of those users never completed the activation event, you have a million trials, not a million customers.

Why it lies: Total user count conflates signups with customers. A signup is a curiosity. An activated user is a customer. The ratio between the two is what matters — and that ratio is often 10–20% for B2B SaaS products.

The test: Divide your total signups by your activated users. If the ratio is above 5:1, your activation rate is below 20% — and you have an activation problem, not a growth problem. The market isn't rejecting your product; your onboarding is.

The Scorecard

The Most Dangerous False Signal

Every founder says this. And it's often true — at the individual level. The CSM who built a relationship with a champion. The founder who personally onboarded the first 20 customers. The support rep who solved every ticket in 2 hours.

Individual love doesn't equal product-market fit. PMF means the product delivers value without the founder, the CSM, or the hero support rep. If your customers love your team but not your product, you have service-market fit, not product-market fit.

The test: Remove the founder and the CSM from the equation. Do customers still stay? Do they still expand? Do they still refer others? If yes, you have PMF. If no, you have relationships — which are valuable but not scalable.

Service-Market Fit vs. Product-Market Fit: A Case Study

A Series A company came to us convinced they had PMF. Their metrics looked great:

• NPS: 72
• TechCrunch feature
• $12M seed round from a top-tier fund
• 400+ "active" customers
• A feature request list 80 items long

But when we pulled the retention data, the picture was different. Their cohort curves decayed to zero. No segment retained above 8% after 90 days. Their $12M wasn't buying PMF — it was subsidizing churn.

Here's what was actually happening: the founder had personally onboarded the first 60 customers. The CSM team (3 people) managed the next 150. Customers stayed because the team was exceptional — not because the product was sticky. Service-market fit: the product is average, but the team around it is extraordinary.

The moment they tried to scale past 400 customers, the retention problem hit them like a wall. CAC was climbing because the "amazing product" story didn't hold up when prospects actually tried it without the founder's involvement. The NPS dropped from 72 to 41 because the CSM-to-customer ratio went from 1:50 to 1:133.

The fix: They narrowed their ICP from "any SaaS company" to "healthcare compliance teams with 20-200 employees" — a segment where the product's specific feature set (audit trails, HIPAA reporting) actually mattered. Within 6 months, retention for that segment flattened at 28%. The other segments still decayed. They had PMF for healthcare compliance teams, not for "SaaS."

The lesson: If your customers would leave if your CSM quit, you don't have PMF. You have service-market fit. The transition from service-market fit to product-market fit requires narrowing your ICP to the segment where the product — not the team — delivers the value.

The PMF Trajectory: What Happens After You Get the First Signs

PMF isn't binary. It's a trajectory. The companies that succeed are the ones that recognize the first sign and then methodically build the others.

The typical sequence:

1. Retention goes flat first. One segment sticks. You didn't plan it — you just notice that customers from a specific industry or company size aren't leaving.
2. Organic referrals follow. The sticky segment tells their peers. You notice more signups from "a colleague recommended it."
3. Pricing power emerges. You raise prices and the sticky segment stays. The non-sticky segment churns — which is actually good news, because it confirms your ICP.
4. Expansion revenue accelerates. The sticky segment starts buying more seats, upgrading plans, or increasing usage. NRR crosses 110%.
5. Sales cycles shorten for the ICP. Prospects who match the sticky segment come in already convinced. Your sales team spends less time persuading and more time confirming fit.

If you're at step 1, don't skip to step 4. Build the evidence chain. Each sign validates the previous one and de-risks the next one.

FAQ