Technical question · Statistics
Explain what a p-value does and doesn’t tell you, to a shop manager deciding whether to keep a new store layout.
What they’re really asking
Two things are tested at once: whether you understand p-values correctly, which many people don’t, and whether you can explain them to someone with a decision to make rather than a statistics exam to pass. The interviewer wants a plain definition tied to the layout, a clear warning about what a p-value can’t tell the manager, and a steer towards what should decide it: how big the effect is, how sure we are of that, and what the layout costs.
How to structure your answer
- Start from the decisionThe manager wants to know whether the layout is worth keeping.
- Say what it tells youIf the layout made no difference, how surprising would a result like this be?
- Say what it doesn’tNot the chance the layout works, not how big the effect is, and not whether it pays.
- Point to what should decide itThe likely size of the gain, its range, the cost, and whether the test was fair.
For more on answering this kind of question, including the mistakes to avoid, read our guide to technical questions.
An example outline
This is an example outline, not a script. Use it to see the shape of a strong answer, then build yours from your own experience, in your own words.
- Decision
- You want to know whether the new layout sells more, and whether that’s enough to roll it out.
- What it tells you
- Say the p-value is 0.03. If the layout made no difference at all, a gap in sales this big or bigger would turn up by chance only about 3 times in 100. So chance alone is an unlikely explanation.
- What it doesn’t
- It isn’t the chance the layout works, and it says nothing about how much it adds. A tiny gain can have a small p-value if we measure enough sales.
- What matters
- The gain is about £400 a week per store, but it could plausibly be anywhere from about £40 to £760. Set that range against the cost of refitting.
- Fair test
- All of this assumes the trial stores were like the others and nothing else changed, such as a promotion. I’d check that before you decide.
Related questions
- A colleague says a rise in click-through rate from 2.0% to 2.3% is “significant”. What would you ask before agreeing?Statistics · 2 minutes to answer
- What’s the difference between correlation and causation? Give an example from real life.Statistics · 1 minute 30 seconds to answer
- A charity wants to test whether a shorter online donation form raises more money. The form gets about 8,000 visitors a month, 5% of them donate, and the average gift is £30. How would you design the test, and how would you decide when to stop it?4 minutes to answer
- A patient survey says 62% of patients rated their GP practice as good, give or take 5 points, and a practice nearby scored 58%. How would you explain to the first practice’s manager what those figures can and can’t tell her?Statistics · 2 minutes to answer
Read the guide
- How to talk through a data project in an interviewInterview guide · 7 min read
- How to explain your SQL out loud in an interviewInterview guide · 7 min read
- How to answer a data pipeline design question as a junior data engineerInterview guide · 7 min read
Last reviewed 28 September 2026 by the Deeplink Interview team. How we write and check our questions.