Product–market fit is the point where customers start pulling your product out of your hands. Before it, every sale feels like pushing a loaded cart uphill. This guide shows you which signals to trust, which ones mislead founders, and five experiments you can run this month to see where you stand.
What product–market fit actually means
Investor Marc Andreessen gave the definition most founders still use: product–market fit means “being in a good market with a product that can satisfy that market.” He also described what its absence feels like. Customers are not quite getting value. Word of mouth is not spreading. Sales cycles drag on and deals do not close.
Michael Seibel of Y Combinator puts it more bluntly. He says product–market fit is what happens after you build what customers want, and the signal is usage that feels explosive. He also warns, “You can’t pretend your way out of product/market fit.”
Keep two points in mind. First, fit is always with a specific group of customers, not “everyone in Pakistan”. Second, fit is not a trophy you win once. A competitor, a price change or a new regulation can weaken it, so you keep measuring.
The signals that matter
Customers come back without being chased
Retention is the strongest signal you have. For a consumer app it means weekly use. For an e-commerce brand it means repeat orders. For a B2B product it means renewals and accounts that grow.
Look at retention by cohort. Group customers by the month they started and track what share are still active one, two and three months later. If every cohort’s curve keeps sliding towards zero, you do not have fit yet. If the curve flattens, even at a modest level, a core group is getting lasting value.
Customers bring other customers
Organic referrals are hard to fake. Listen for “my cousin told me” or “someone forwarded your number in our WhatsApp group”. Ask every new customer how they heard about you and record the answer in a sheet. A rising share of word-of-mouth customers is a strong sign.
Customers pay, and pay on time
In Pakistan, payment behaviour tells you a lot. If you sell with cash on delivery, track how many orders are refused or returned at the door. If you sell to businesses, track how many days invoices take to clear. Customers who pay full price, on time, without being chased are telling you the product matters to them.
Customers complain when it breaks
If your app goes down on a Saturday and nobody messages you, that is a warning. Angry messages about an outage are uncomfortable, but they show people depend on you.
Sales get easier
Watch the time from first conversation to first payment. If it is shrinking and fewer deals stall, your pitch and your product are matching a real need.
Signals that fool founders
These feel like progress but often are not:
- Compliments, likes and follower counts.
- A spike in sign-ups after a press mention or a viral reel, followed by silence.
- Sign-ups that depend on discounts or free trials and never convert to paid.
- Pilots with large companies that never become paid contracts.
- Growth that comes only from ad spend and stops when the spend stops.
Seibel makes a related point about growth on its own. He describes an earlier company of his that was profitable but still lacked fit, because it had no repeatable engine for growth.
Five experiments that show where you stand
1. The 40% survey
This test comes from Sean Ellis, who led early growth at Dropbox, LogMeIn and Eventbrite. The email company Superhuman made it famous. Superhuman surveyed users who had used the product at least twice in the previous two weeks. The key question asked how they would feel if they could no longer use Superhuman. The answers were very disappointed, somewhat disappointed and not disappointed. For your version, put your own product’s name in the question.
Ellis found that companies where fewer than 40% of users said “very disappointed” usually struggled to grow. Superhuman added three follow-up questions: what type of people would most benefit, what is the main benefit you receive, and how can we improve the product for you. Its score rose from 22% to 58% within three quarters.
To run it, send a short form to active users only, by WhatsApp or email. Offer it in English and Urdu if your customers prefer Urdu. Do not read too much into a handful of answers. Wait until you have enough responses to see a pattern.
2. The retention cohort sheet
Build a simple table in Google Sheets. Each row is the month customers joined. Each column is the share still active in later months.
| Cohort (hypothetical) | Month 1 | Month 2 | Month 3 | Month 4 |
|---|---|---|---|---|
| January | 100% | 42% | 35% | 34% |
| February | 100% | 45% | 38% | 37% |
| March | 100% | 51% | 44% | — |
This example is made up for illustration. The pattern to look for is the flattening after month two, and newer cohorts doing slightly better than older ones.
3. The segment cut
Superhuman did not average all its answers. It looked closely at the users who said “very disappointed”, described who they were, and built for them. Narrowing its focus this way raised its score by about ten points.
Do the same. Sort your survey answers and retention data by customer type, city, order size or use case. A hypothetical example: a tutoring app in Lahore finds that its “very disappointed” users are mostly parents of O-level students. That is a segment worth serving deeply before expanding.
4. The price test
For one or two weeks, show new customers a higher price, or remove a standing discount. If conversion barely moves, customers value the product more than you assumed. You can also test commitment by asking a small group to pay in advance by bank transfer or Raast instead of cash on delivery, and see how many accept.
5. The referral ask
Ask your ten most active customers for one introduction each to someone with the same problem. Count how many follow through within a week. An introduction costs the customer some reputation, so it is a stronger signal than a compliment.
What to do with the results
Superhuman split its roadmap in two. Half the time went to doubling down on what its happiest users loved. The other half went to fixing what held back users who liked the main benefit but were not yet hooked. As its founder put it, “If you only double down on what users love, your product-market fit score won’t increase.”
If the numbers stay weak after several rounds, change one big thing at a time: the segment, the problem, or the offer. Seibel’s advice for teams that have not found fit is to talk to users more, and to cut spending so you can afford to keep learning.
Your product–market fit checklist
- Write your target customer in one sentence, including city or segment.
- Set up a monthly retention cohort sheet and update it on the first of every month.
- Run the 40% survey with active users only, then sort answers by segment.
- Ask every new customer how they heard about you and log the answer.
- Track payment behaviour: cash-on-delivery refusals, returns and days to pay.
- Run one experiment at a time, for a fixed period, and write down what you expect before you start.
- Review the results with your co-founders each month and decide what to double down on and what to fix.
Sources and further reading
Comments