Analyze Profile Traffic Drop
Company: LinkedIn
Role: Data Scientist
Category: Analytics & Experimentation
Difficulty: hard
Interview Round: Technical Screen
LinkedIn observes a noticeable decline in traffic from the Home Page to the Profile Page. You are asked to investigate whether this decline is caused by a product bug, a logging issue, reduced user interest, or a positive product change that reduces the need to click through.
How would you analyze this problem end to end?
In your answer, address:
- How you would define the problem precisely (for example, absolute Profile Page visits, click-through rate from Home to Profile, unique viewers, or visits per Home session).
- What top-line metrics you would check first, such as total Home Page traffic, total site traffic, session length, and downstream engagement.
- How you would verify instrumentation and rule out data-quality issues.
- How you would break the problem down by platform, country, member segment, referrer, and traffic path.
- How you would identify where users went instead if they no longer clicked into Profile Pages.
- How you would distinguish a harmful drop from a healthy UX improvement.
Assume one possible recent product change is that users can hover over a member's name on the Home Page to preview profile information without opening the full Profile Page. Explain how you would test whether this feature is responsible for the decline.
Also answer this follow-up: if total user time spent on LinkedIn stays flat while Home-to-Profile visits decline, how would that change your interpretation?
Quick Answer: This question evaluates a data scientist's competency in product-metrics analysis, instrumentation validation, segmentation and attribution, and causal inference for diagnosing traffic and engagement changes, and is commonly asked to assess analytical design, experimental thinking, and cross-functional reasoning.