Interview scorecard
Data Analyst
Score each skill from 1 to 5. A 1 is below the bar for this role, a 3 meets it, and a 5 is above it. The examples show what each sounds like. Use 2 and 4 for answers that fall in between.
SQL
Writes accurate, readable queries that pull the right data from several tables and checks the results.
Questions
- Tell me about a complex query you wrote. What was it answering and how did you build it?
- Describe a time a query ran slowly or returned odd numbers. What did you do?
- Tell me about a time you joined several tables. How did you know the row counts were right?
1: Below the bar
Describes only simple single-table queries. Cannot explain joins or how they would spot wrong or duplicated results.
3: Meets the bar
Builds multi-table queries with joins and grouping. Checks results against a known number and fixes slow queries with some help.
5: Above the bar
Explains joins, filters and window functions clearly. Tests results at each step, finds the cause of slow or wrong output, and shares fixes.
Data analysis
Turns a vague business question into a clear, tested answer drawn from messy data.
Questions
- Tell me about an analysis where the question was vague at first. How did you narrow it?
- Describe a time your data showed something surprising. How did you check it was real?
- Tell me about a time your first answer was wrong. How did you find out?
1: Below the bar
Runs numbers without defining the question. Accepts surprising results at face value and can't say how they'd check them.
3: Meets the bar
Frames the question with some prompting. Checks odd results and names one or two other explanations, but stops early.
5: Above the bar
Sharpens the question first, tests several explanations unprompted, and says how confident they are and what could change the answer.
Data storytelling
Turns analysis results into a clear message that helps non-technical people decide what to do.
Questions
- Tell me about a time you presented findings to people who don't work with data.
- Describe a chart or report you changed because your audience found it confusing.
- Tell me about a time your findings led someone to act or decide differently.
1: Below the bar
Lists numbers and charts without a main point. Doesn't adjust the language or detail for the audience.
3: Meets the bar
States a clear main finding and picks suitable charts. Adjusts language for the audience but gives only a vague recommendation.
5: Above the bar
Leads with the point, cuts detail the audience doesn't need, and ends with a clear recommendation. Gives an example of a decision it changed.
Business understanding
Connects analysis to how the company makes money and what decisions the business needs to make.
Questions
- Tell me about a time you asked what decision an analysis request was really for.
- How would you explain what drives revenue or costs at your last company?
- Tell me about a time your work changed a business decision. What happened?
1: Below the bar
Takes requests literally and can't say how the business earns money. Describes outputs, not the decisions they supported.
3: Meets the bar
Asks what a request is for and explains the main business drivers. Links some past work to a decision.
5: Above the bar
Reframes requests around the decision at stake, explains drivers with numbers, and shows a decision or result their work changed.
Attention to detail
Checks data and results for errors before sharing them with others.
Questions
- Tell me about a time you caught an error in your own analysis. How?
- What checks do you run on a new dataset before trusting it?
- Tell me about a time a small mistake in your work reached someone else. What happened?
1: Below the bar
Trusts data and results as they come. Mentions no checks, or finds errors only when someone else points them out.
3: Meets the bar
Runs routine checks like row counts and missing values. Can describe an error they caught and fixed before sharing.
5: Above the bar
Has a repeatable checking habit, reconciles results against another source, and explains how past slips changed their process.
Overall recommendation
Decide on your own, before you compare notes with the rest of the panel.