BY JOSHUA AJIEH
Weak growth alone cannot tell you whether you have a product problem or a distribution problem. Investigate whether suitable customers discover the product, understand its promise, reach a useful outcome and return when they need it again.
Treat each stage as a separate question. The aim is to find the constraint worth testing, rather than choose a label for the whole business.
Define the two questions.
For this investigation, distribution concerns how you reach suitable customers and communicate an offer they understand. Product value concerns whether the experience helps those customers accomplish the task that brought them in.
The boundary is not always clean. A promise can attract an unsuitable audience. A signup flow can prevent an interested visitor from trying the product. A useful product can serve a need that arises only occasionally.
Start with a specific customer and problem.
Before analysing traffic, describe who the product is intended to help, the situation that creates their need and the outcome they want.
- Customer: Who faces this problem?
- Situation: What happens that makes them look for a solution?
- Outcome: What would a useful result look like to them?
- Alternative: How do they handle the problem today?
Use observed behaviour and customer conversations to refine that description. Someone matching your intended customer profile is a candidate to investigate, not proof of demand.
Locate the break in the journey.
Use the patterns below to choose an investigation. They are starting hypotheses, not diagnoses.
| Observed pattern | Question to investigate | Evidence to collect |
|---|---|---|
| Few suitable people discover the product | Are you reaching the intended audience in a place where they look for help? | Acquisition sources, audience characteristics and conversations about how customers find solutions. |
| People visit but do not sign up | Do the audience, promise and signup commitment match? | Visitor feedback, message comprehension, CTA clicks and signup completion. |
| People sign up but do not reach value | What prevents the first useful outcome? | First-task completion, errors, time to outcome and observed attempts to use the product. |
| People reach value but do not return | Was the result useful enough, and when does the need arise again? | Repeat meaningful actions, expected usage frequency and interviews with returning and non-returning users. |
| People use the product but do not pay | Who receives value, who pays and what makes the offer worth purchasing? | Purchase attempts, buyer conversations, pricing feedback and the alternatives customers pay for. |
Compare outcomes by audience and source.
An overall conversion rate can hide differences between groups. Compare what happens after visitors arrive from each source, and whether those visitors have the problem you intend to solve.
- Which sources bring people with a relevant need?
- What promise did each group see before arriving?
- Who completes the first meaningful action?
- Who returns when the need arises again?
- Who attempts to purchase, and where do they stop?
Compare groups over appropriate time periods and show the underlying counts alongside percentages. A high rate from a handful of users is a clue to investigate, not enough evidence to scale a channel confidently.
Check the outcome behind the activity.
Account creation, clicks and app opens show activity. Ask whether users accomplished something useful. Observe them attempting the task and ask what they did with the result.
If people reach an outcome but do not return, investigate the next occasion when they face the same problem. Did they use your product, choose an alternative or no longer need a solution?
Read the guide to investigating why users try your product once and do not return.
Do not confuse encouragement with demand.
A customer saying “this looks useful” does not show what they will do when the problem occurs. Ask about recent behaviour, existing workarounds and the effort or money they already commit to solving it.
When appropriate, test a clear offer with an observable commitment. State what the customer will receive and what the commitment involves.
Explore the difference between customer praise and evidence of product demand.
Run one experiment you can interpret.
Choose the explanation that best fits the evidence you have. Write down what you would expect to observe if it were true, then design a focused test.
For example, imagine that visitors cannot explain what the product helps them accomplish. You could test a clearer promise with a comparable audience while keeping the offer and signup flow unchanged. This is an illustrative test, not a diagnosis of your website.
Alternatively, if users understand the offer but become stuck during their first task, test a change to that step while keeping acquisition conditions as consistent as practical.
- Hypothesis: What do you think is preventing progress?
- Audience: Who will participate, and why are they relevant?
- Change: What will you test?
- Measure: Which behaviour or outcome will you observe?
- Decision: What result would support, weaken or leave the explanation unresolved?
Avoid changing the audience, message, price and product journey simultaneously if your aim is to understand which change mattered.
Choose the next move from the evidence.
If suitable users repeatedly get value but too few discover the product, test ways to reach more people with similar needs. If suitable users struggle to reach value, investigate the experience before committing to more acquisition spending.
If both stages show problems, choose a focused sequence of tests. You do not need to declare the business entirely a product problem or entirely a distribution problem.
Work through your product constraint.
Stop Building, Start Validating provides a practical framework for questioning assumptions and choosing what to test next. For help examining a live challenge, explore Product Diagnosis.