Product design question

How would you improve Google Photos?

Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.

What this question tests

Product sense for a mature product: finding a focused, high-impact improvement rather than a laundry list.

Pick a specific user segment, like people with years of photos who struggle to find a specific memory, rather than 'all users.'
Identify the core pain point: search still misses many photos because of weak captioning or ambiguous queries.
Prioritize improving natural-language search over adding new editing features, since findability is the product's core promise.
Propose a concrete improvement: richer AI-generated captions combined with conversational search that handles vague queries like 'that trip with the dog.'
Define success metrics: search success rate (click-through on results) and reduction in zero-result searches.

Names a specific gap, like vague or multi-object queries failing today, instead of a vague 'better AI' pitch.
Proposes a concrete mechanism, like combining visual embeddings with conversational query understanding.
Gives an illustrative metric, e.g. cutting zero-result searches from 15% to 5%, marked as an assumption.
Considers the trade-off of compute cost for richer captioning against expected search engagement gains.

Listing unrelated features (editing, sharing, storage) instead of going deep on one clear improvement.
Ignoring that search quality is Google Photos' core differentiator versus competitors.

Chapters of the AI PM course that teach what this question tests.