Data Analyst interview question
How do you design a useful dashboard?
A useful dashboard serves one decision and audience, leads with a small set of defined metrics, shows target and trend context, supports intentional drill-down, states freshness, and avoids decorative charts without action.
Direct answer
What a strong answer covers
How it works
A dashboard exists to support a specific recurring decision for a specific audience, and everything on it is judged by whether it helps that decision. The top row shows the few metrics that matter with their targets and recent trend, so a glance answers whether things are on track. Below it, breakdowns that explain movements in those metrics, in a drill-down order that follows how the audience investigates. Every number has a definition on hover or a link, the data freshness is displayed, and charts use the form that fits the comparison: lines for trends, bars for categories, tables for exact values. Anything decorative or unactionable is removed.
Example
A weekly growth review dashboard leads with new sign-ups, activation rate, and week-4 retention against targets, each with a 12-week sparkline; the second row breaks sign-ups by channel and activation by platform because those are the two levers the team controls. A pie chart of browser share was removed after nobody could name a decision it informed.
Common mistakes
- Showing thirty tiles with no hierarchy so nothing stands out.
- Presenting metrics without target or trend, leaving viewers unable to judge good or bad.
- Omitting the last-refresh time and letting stale data drive a decision.
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.