Metrics question
How does the mean and median compare for search response times in Google?
- Metrics
- Medium
Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.
Start a mock interview on this question · Mock interview from a job description
What this question tests
Statistical reasoning applied to a real metrics question, checking you understand what mean versus median reveals about a distribution.
How to approach it
- Clarify what response time is being measured, likely server-side query latency for search results, and over what population of queries.
- State the general relationship: for a right-skewed latency distribution, mean is typically higher than median because a small number of slow outlier queries pull the average up.
- Explain what causes the skew specifically for search, such as complex queries, cold cache hits, or geographic distance from serving infrastructure.
- Discuss why this matters for product decisions, since median reflects typical user experience while mean can be dominated by rare slow requests.
- Recommend using percentile metrics like p95 or p99 alongside median, since those better capture the tail experience that mean and median alone can hide.
- Confirm whether the interviewer wants the general statistical relationship or a discussion of why Google's specific latency might behave this way.
What a strong answer includes
- States clearly that mean exceeds median in a right-skewed latency distribution and explains the causal mechanism, outlier slow queries, not just the fact.
- Names specific causes of tail latency for search, like cache misses or complex query types, grounding the answer in a real system.
- Proposes percentile-based metrics, p95 or p99, as more actionable than either the mean or median alone for catching tail-latency problems.
- Connects the statistical point to a product consequence, since a small number of very slow queries can disproportionately hurt perceived quality even if the median stays low.
Common mistakes
- Not explaining why mean and median diverge, just stating that they do.
- Ignoring percentile metrics entirely, missing the more actionable insight for latency work.
- Confusing the direction of the skew, claiming median would be higher than mean for a typical latency distribution.
Likely follow-up questions
- Why would p99 latency matter more than mean latency for user experience?
- What could cause the median itself to increase even if p99 stays flat?
More metrics questions
- How do you define success for Yelp reviews?Google · Metrics · Medium
- Utilization went down by 45% on app XYZ in Italy for the month of August. Give a reason why and draft a plan to fix it.Spotify · Metrics · Medium
- You launched a new signup flow to encourage new users to add more profile information. A/B test results indicate that the % of people that added more information increased by 8%. However, 7 day retention decreased by 2%. What do you do?Google · Metrics · Hard
- Define the metrics for YouTube search.Google · Metrics · Medium
- You walk in to your office and find that Google cloud subscription has gone down by 20%. What do you do as a product manager?Google · Metrics · Medium
- What would be the top six metrics for WhatsApp? Now pick one from the six. Tell me three things you would do to improve that metric.Google · Metrics · Medium
More questions from Google
Learn the skill behind it
Chapters of the AI PM course that teach what this question tests.
- Chapter 9: Prove it paid off: outcomes, economics, and pricing
- Chapter 2: Data fluency: SQL, logs, and reading the truth yourself
- Chapter 14: Get the job: the AI PM interview loop