Product-Market Fit Diagnostic
Assess how strongly a product fits real customer needs using multiple converging signals, not a single score.
14 min · Popularised by Marc Andreessen; diagnostic signals widely used across the start-up ecosystem
What is this framework?
Product-market fit means your product satisfies real market demand strongly enough that customers use it, return to it, pay for it, and tell others about it. This diagnostic looks at several signals together — need strength, usage, retention, repeat purchase, willingness to pay, referrals and dependence — to judge how close you are.
Product-market fit (PMF) is one of the most-cited but least-precisely-defined ideas in entrepreneurship. Marc Andreessen described it as being 'in a good market with a product that can satisfy that market' — useful as intuition, but hard to act on directly. This diagnostic breaks that intuition into observable signals a founder can actually check.
Rather than asking 'do we have PMF, yes or no?', the diagnostic asks a set of narrower questions: how strong is the underlying customer need, how often do people actually use the product, do they stay, do they buy again, would they pay a fair price, do they refer others, and would they be upset if the product disappeared? Strong signals across several of these areas suggest genuine fit; weak or mixed signals suggest more iteration is needed.
It's important to treat this as a diagnostic conversation rather than a scorecard with a pass mark. Two ventures with identical 'scores' can be in very different situations depending on their market size, business model and stage. The value of the framework lies in surfacing which specific dimension is weakest, so the team knows what to work on next.
What problem does it help solve?
- Move beyond gut feeling when judging whether a product is ready to scale
- Identify which specific dimension of fit (need, usage, retention, payment, referral) is weakest
- Avoid scaling spend prematurely on a product that isn't yet retained or repurchased
- Create a shared vocabulary for the team to discuss fit honestly
The framework
Need strength
How painful or important is the problem being solved?
Usage
Do customers actually use the product regularly?
Retention
Do they keep using it over time rather than churning?
Repeat purchase
Do they buy again without being pushed?
Willingness to pay
Will they pay a price that sustains the business?
Referrals
Do they recommend it to others unprompted?
Customer dependence
Would they be genuinely disappointed if it disappeared?
These pillars are diagnostic signals, not components of a single formula — read them together, not in isolation.
Every part explained
Need strength
How urgent, frequent or painful the underlying problem is for the target customer.
Ask: How strongly does this problem hurt, and how often does it occur?
Example: Parents needing reliable, safe transport for children to and from tuition classes daily.
Usage
Whether customers actually engage with the product in practice, not just in surveys.
Ask: Are people actually using this, and how often?
Example: 70% of registered users open the app at least three times per week.
Retention
Whether users continue using the product over weeks and months rather than trying it once.
Ask: What proportion of users are still active after one, three and six months?
Example: Cohort analysis shows 60% of month-one users are still active in month three.
Repeat purchase
Whether customers buy again without heavy discounting or reminders.
Ask: Do customers return to buy again on their own initiative?
Example: 45% of first-time customers place a second order within a month.
Willingness to pay
Whether customers will pay a price that can sustain the business, not just a heavily discounted trial price.
Ask: Would customers pay full price for this, and does that price cover our costs?
Example: Customers agree to a RM25 monthly subscription after a free trial ends.
Referrals
Whether satisfied customers proactively recommend the product to others.
Ask: Are customers telling others about us without being incentivised?
Example: A third of new sign-ups arrive via word-of-mouth referral codes.
Customer dependence
How disappointed customers would be if the product suddenly disappeared.
Ask: If we shut down tomorrow, how many customers would genuinely miss us?
Example: In a survey, 40% of users say they would be 'very disappointed' if the product were gone.
Worked example — A tuition-centre scheduling app for Malaysian parents
The founding team runs their six-month-old product through the diagnostic to decide whether to raise funding for scaling.
Need strength
High — parents describe scheduling clashes as a recurring weekly frustration.
Usage
Moderate — 55% of registered parents log in weekly during the school term.
Retention
Weak — only 30% of users from month one are still active by month three.
Repeat purchase
Not yet applicable — the app is currently free, so there is no repeat purchase signal.
Willingness to pay
Untested — the team has not yet asked users to pay.
Referrals
Moderate — some parents share the app in class WhatsApp groups unprompted.
Customer dependence
Mixed — some parents say they'd miss it, others say they'd revert to manual scheduling easily.
The signals point to real need but weak retention and untested monetisation — the team decides to fix the retention problem before testing pricing or raising funds.
How to use it
- 1Gather usage data: logins, active sessions, and drop-off points over time.
- 2Run cohort retention analysis for at least three consecutive time periods.
- 3Ask customers directly how disappointed they would be if the product disappeared.
- 4Test willingness to pay with a real price, not a hypothetical survey question.
- 5Track whether referrals are happening organically, without heavy incentives.
- 6Rate each of the seven signals honestly as strong, moderate or weak.
- 7Identify the weakest one or two signals and prioritise fixing those before scaling.
Try it yourself
Rate your own product against each signal and note the evidence behind your rating.
Diagnostic signals
Your work stays on this device. Nothing is uploaded, so use the same browser to come back to it.
When to use it
- After launching a minimum viable product and gathering a few months of usage data
- Before deciding whether to invest heavily in scaling or paid acquisition
- When preparing a fundraising narrative and needing honest evidence, not just optimism
When not to rely on it
This framework does not prove:
- • This is not a single universal score — it does not mathematically certify product-market fit
- • Signals can look strong in a small, unrepresentative early-adopter group
- • Different business models (e.g. one-off purchases versus subscriptions) need different signal weightings
Common mistakes
- Declaring 'we have PMF' based on one strong signal (e.g. enthusiastic reviews) while ignoring weak retention
- Testing willingness to pay with hypothetical questions instead of real transactions
- Measuring usage only in the first week and ignoring longer-term cohorts
- Assuming referrals prove fit when they were driven entirely by cash incentives
Connections
Related concepts
Quick check
Why does the product-market fit diagnostic use seven separate signals instead of one score?
Remember this
Product-market fit isn't a single milestone you hit — it's a pattern across several honest signals, and the diagnostic's job is to show you which one needs attention next.