What is product-market fit?
Product-market fit is a relationship between a product, a particular customer group, and a valuable job those customers need done. It becomes credible when people repeatedly choose the product, obtain the intended result, and keep using or buying it under realistic conditions. A product can fit one narrow segment while disappointing another.
For a small SaaS, start with a precise claim: ‘Independent bookkeeping firms use this every month to collect missing client documents.’ That is easier to evaluate than ‘Small businesses love our platform.’ Ask what behavior would support the claim, what would contradict it, and how long a customer needs to experience the recurring job.
Idea validation happens earlier. Interviews and a paid pilot can show that a problem deserves work, but they cannot yet establish reliable long-term retention. If you have no usable product or active users, begin with validating a micro-SaaS idea, then return to this review when customers have experienced the result.
Use the product-market fit survey as a diagnostic
In his first-person account of Superhuman, Rahul Vohra describes using Sean Ellis’s disappointment question, then segmenting responses and changing the roadmap. The often-cited benchmark is 40% choosing the strongest disappointment response. Treat it as a practical heuristic from that method, rather than a universal pass mark or proof that your business can scale.
Keep the core question and answer options stable if you want to compare waves. Our downloadable template adds original questions about the completed job, alternatives, and obstacles. Those additions are for learning; they are not a validated scoring instrument. Avoid turning every answer into a numeric ‘PMF index.’
How would you feel if you could no longer use [product]? • Very disappointed • Somewhat disappointed • Not disappointed
Choose respondents who have experienced the core value
Define eligibility before sending the survey. Include people who have had a fair opportunity to complete the main job, then record their segment, product version, and observation period. For the fictional bookkeeping tool, you might invite firms that completed at least one client-document collection cycle in the last 60 days. That is an example criterion to adapt to the product’s actual cadence.
Someone who signed up yesterday and never imported a client cannot assess the recurring outcome yet. Keep those activation problems in a separate study. Also review people who recently stopped using the product: surveying only current enthusiasts can hide why the broader group leaves. Do not merge churned-user interviews into the active-user score without changing and labeling the population.
Record how many eligible users were invited, how many responded, and how respondents differ from nonrespondents. Ask each person once per defined wave, deduplicate responses, and make follow-up contact optional. Do not ask for client documents or other sensitive records just to understand the workflow.
An original follow-up survey you can adapt
Put the short disappointment question first, then ask about an actual recent experience. Questions about the last completed job produce more useful detail than asking whether a hypothetical feature sounds good. Keep the invitation neutral and make it clear that critical feedback is welcome.
The following questions are our original learning template. Send only the ones needed for the decision you are making. If you already know the customer’s role from an appropriate account record, avoid asking them to repeat it.
- What job were you trying to complete the last time you used the product?
- What result did you get, and what remained unfinished?
- What would you use or do instead if this product disappeared?
- Which part of the workflow saves you the most effort, and why?
- What nearly stopped you from completing that job?
- Who else has this same problem, and how is their situation similar?
- Optional: may we contact you about this response?
Calculate the score without hiding a small sample
Survey score = eligible respondents selecting the strongest disappointment response ÷ eligible respondents answering the core question × 100. Keep skipped questions out of that denominator and show the count of skips separately. The number describes respondents, not every customer or everyone in the market.
Here is a fictional survey: 80 eligible accounts are invited, 20 answer the core question, and nine choose the strongest response. The score is 9 ÷ 20 × 100 = 45%, with a 25% response rate. One different answer would move the score by five percentage points. The score crosses the familiar heuristic, but 60 nonrespondents remain an important uncertainty.
Break down a preselected segment if the question is whether a particular audience gets value. Do not search through dozens of tiny slices until one looks impressive and present that slice as the whole market. Preserve the original population and totals when explaining a narrower opportunity.
| Observation | Count | Meaning |
|---|---|---|
| Eligible accounts invited | 80 | The defined population contacted |
| Core question answered | 20 | 25% response rate |
| Strongest disappointment response | 9 | 45% of respondents |
| Nonrespondents | 60 | Their views are not established by this survey |
Triangulate survey responses with customer behavior
A useful evidence review asks whether the reported benefit appears in behavior. Are eligible customers completing the core job again at its natural frequency? Do paying accounts renew? Can the team deliver the result without an unrecorded founder rescue every time? Look at customers old enough to have reached the relevant renewal or usage window.
Separate a strong product experience from a sustainable business model. A product may be loved at a free price while the cost of serving customers exceeds what they will pay. Conversely, a few prepaid annual contracts do not yet tell you what will happen at renewal. Keep retention and churn, pricing feedback, and delivery effort visible together.
| Signal | Evidence to collect | Common trap |
|---|---|---|
| Repeated value | Completed jobs across mature cohorts | Counting logins instead of outcomes |
| Demand | Paid renewals, referrals, and reasons for buying | Treating a launch spike as recurring demand |
| Retention | Opening cohorts with completed observation windows | Mixing new signups into the denominator |
| Economics | Delivery costs and actual acquisition effort | Ignoring founder time and manual support |
Turn the review into a focused product decision
Write down the audience receiving the clearest benefit, the task they rely on, and the most common obstacle to repeating it. Choose one change that should make that task more dependable. For the fictional document tool, this might be clearer missing-file requests rather than a new dashboard, because customers report chasing the same unclear requests every cycle.
Set a follow-up window long enough to observe the job again. Review whether the change improved completion and repeat usage, then survey the next eligible wave using the same rules. Keep earlier results and population definitions so a different audience mix is not mistaken for product improvement. Use the customer interview guide to investigate contradictory responses.
Download the product-market fit survey template
The editable template includes eligibility, a reference to the attributed core method, an explicitly labeled original question adaptation, our follow-ups, a response-count worksheet, and a decision log. Keep the original question unchanged when tracking the original method; changed wording may produce different responses. Source checked October 26, 2026.