Product Analyst Interview Questions UK: 8 Real Questions and What Interviewers Want

Applying for a Product Analyst job in the UK? Here is what you are likely to be asked. These questions come from our Product Analyst question bank of 16 questions, written for UK employers at mid-level level.

Eight Product Analyst interview questions, and what the interviewer is thinking

1. How would you design and analyse an A/B test for a new checkout flow, from hypothesis to decision?

Why they ask: Experimentation is the core skill of most product analyst roles.

2. How would you build a funnel and retention analysis for a mobile app, and what tools would you use?

Why they ask: Funnels and retention are everyday product analysis work.

3. What is a north star metric, and how would you choose one and supporting metrics for a product?

Why they ask: Metric framing shows product thinking, not only technical skill.

4. Daily active users have fallen by ten per cent over the last week. How would you investigate?

Why they ask: Diagnosing metric movements is a standard product analyst scenario.

5. Write or describe a SQL query to calculate week-one retention by sign-up week.

Why they ask: SQL for product metrics is tested in most analyst interviews.

6. What is the difference between statistical and practical significance, and how do you deal with multiple testing and segment slicing in experiments?

Why they ask: Strong candidates avoid common experimentation pitfalls.

7. Tell me about a time your analysis influenced a product roadmap or feature decision.

Why they ask: To find evidence of real impact and the ability to influence stakeholders with data.

8. What does a typical week look like for a product analyst, and how do you work with product managers?

Why they ask: To check that you understand the day-to-day and partnership nature of the role.

A sample answer to question 1

How would you design and analyse an A/B test for a new checkout flow, from hypothesis to decision?

I would begin with a clear hypothesis, such as 'a single-page checkout will increase completed purchases', and pick one primary metric, checkout conversion, plus guardrail metrics such as refund rate, page load time and average order value. Next I would run a power calculation to determine the sample size needed, based on the baseline conversion rate, the minimum detectable effect that would justify the build cost, a significance level of 0.05 and power of 80 per cent, which tells me how long to run it. I would randomise at user level, check there is no sample ratio mismatch, and run for full weekly cycles to cover day-of-week effects. I would avoid peeking and stopping early unless using a sequential method. At the end, I would report the effect with a confidence interval, check segments for sensible differences without over-slicing, and recommend ship, iterate or stop, noting any risks such as novelty effects.

Common mistake to avoid: Stopping when the p-value first dips below 0.05; ignoring guardrail metrics

Want the rest?

Full model answers, key points and common mistakes for all 16 questions are on the Product Analyst interview questions page. The first five are free, no sign-up needed.