Data Analysis Questions
Part of the Business Analyst Roadmap
In short
'The marketing team says the new campaign doubled sign-ups. How would you check?' These questions test whether you trust data blindly or question it like an analyst. Bonus points for a quick SQL query.
'How would you validate this business claim with data?' — testing Phase 5's data analysis judgment and basic SQL/Excel fluency against a realistic scenario.
Think of it like this
Checking the 'tough paper' claim
The class says everyone failed because the paper was tough. A sharp teacher checks: did last year's batch score better on similar papers? Did only one section fail? Did attendance drop? Claims become facts only after you check the data.
Watch it happen
How to answer 'validate this claim'
1/5Define the claim exactly: which sign-ups, which dates, and doubled compared to what?
The big ideas
What you need to know
Define before you measure
'Sign-ups' might mean accounts created or accounts verified. Pin it down first.
Compare to a baseline
A number means little alone. Compare against the period before, or a group without the change.
Correlation isn't cause
Two things rising together doesn't prove one caused the other. Name other possible causes.
Show basic SQL or Excel
Writing a simple GROUP BY or pivot on the spot shows you can actually do it, not just talk.
See it in code
Walk through the code
1SELECT source,2 COUNT(*) FILTER (WHERE created_at < '2026-09-15') AS before,3 COUNT(*) FILTER (WHERE created_at >= '2026-09-15') AS after4FROM signups5WHERE created_at >= '2026-09-01'6 AND created_at < '2026-09-29'7 AND is_verified = true8GROUP BY source;1/4Count sign-ups before and after launch for each source.
Two weeks before and after a campaign launched on 15 September, split by source, verified accounts only. Written for PostgreSQL.
Save yourself the pain
Common mistakes
Oops
Accepting the claim and moving to 'how we can scale it'.
Do this instead
Validate first. Show healthy scepticism, then talk about next steps.
Oops
Freezing when asked to write a query.
Do this instead
Talk through the logic in plain words first, then write it. Interviewers value the reasoning.
Your mission
Validate a claim yourself
Quick check
Sign-ups doubled the week a campaign launched. What should you check before crediting the campaign?
Go deeper
- 1Review Phase 5 and practice framing a business question as a specific data question.
- 2Prepare to discuss a real analysis you've done, even from a practice exercise.
- 3Practice explaining a basic SQL query or Excel technique you're comfortable with.
More process & analysis at this stop
Process Modeling Questions
Gap Analysis Scenarios
Stuck on this topic? Ask an Insider
Get 1:1 guidance from people who've walked this exact path — on the InsideEdge app.
