Full-stack Developer interview question
How do you diagnose a slow user journey?
Trace the complete user journey before optimizing: measure browser rendering and network waterfalls, follow the request through services and queries, identify the dominant latency component, fix it, and compare the same percentile baseline.
Direct answer
What a strong answer covers
How it works
A slow journey is diagnosed from the outside in. Field data from real users identifies which page or action is slow and for whom. The browser waterfall shows whether time goes to blocking resources, a long-running API call, or client-side rendering after the data arrives. The trace for that API call shows which service, query, or external dependency dominates. Only then is a fix chosen, and it is judged against the same percentile on the same segment, because a median improvement can hide a p95 that got worse.
Example
Checkout felt slow on mobile; field data showed p95 of 6 s on 4G. The waterfall showed the page waiting 3.5 s on a single API call, and the trace showed that call fanning out to a tax service with no timeout. Adding a 500 ms timeout with a cached fallback rate brought p95 to 2.1 s, while the bundle-size work that was first proposed would have saved 200 ms.
Common mistakes
- Starting with bundle optimisation before knowing where the time goes.
- Measuring on a developer machine with a warm cache and a fast connection.
- Reporting the average and missing the slow tail that drives complaints.
A useful answer structure
Answer from your own work.
Do not memorise a script: use only examples you can explain honestly in follow-up questions.
Open the desktop workspace and rehearse a clear answer in your own words.