Role interview guide
Data analyst interview questions and what to expect
Data analyst interviews usually test three things: whether you can get the right numbers out of data (SQL and spreadsheets), whether you can reason about what those numbers mean for the business, and whether you can explain a finding to someone who isn't technical. Many companies add a take-home or live exercise on a small dataset.
What candidates commonly report; it varies by employer, level and country. Last reviewed 18 October 2026.
The usual stages
- 1Recruiter or HR screen
- 2Online test or live exercise: SQL, spreadsheet work and basic statistics
- 3Case or take-home: analyse a small dataset and present what you found (often)
- 4Hiring manager interview on past projects and business sense
- 5Final or HR round
What interviewers check
- Accurate SQL: joins, grouping, window functions, handling nulls and duplicates
- Turning a vague business question into a metric you can measure
- Checking data before trusting it
- Clear charts and a one-line takeaway
- Explaining limits and uncertainty honestly
Behavioral questions
Tell me about an analysis you did that changed a decision.
A good answer covers
- The decision at stake and who was making it
- What you analysed and the one finding that mattered
- What changed because of it, with a number if you have one
Tell me about a time you found a mistake in a report or dataset others were relying on.
A good answer covers
- How you noticed something was off
- How you confirmed it before raising it
- How you told people and what you fixed so it didn't recur
Describe a time a stakeholder disagreed with what your data showed.
A good answer covers
- Their view and why it was reasonable to them
- How you checked your own numbers again
- How you reached a shared answer, or agreed what to test next
Tell me about a time you had to deliver an analysis with a very tight deadline.
A good answer covers
- What you cut and what you kept
- How you made the limits of a quick answer clear
- What you followed up with afterwards
Questions about the job
Daily orders dropped 15% yesterday. How would you find out why?
A good answer covers
- Checking the data first: a tracking or pipeline problem, a holiday, a one-off
- Breaking the drop down by city, platform, channel and new versus returning users
- Stating the most likely cause and how you'd confirm it
How would you find the top three products by revenue in each category using SQL? Talk me through it.
A good answer covers
- Aggregating revenue per product, then ranking within each category
- A window function such as ROW_NUMBER or DENSE_RANK, and how ties are treated
- Filtering to the top three after ranking, not before
What's the difference between an inner join and a left join, and when has the choice mattered in your work?
A good answer covers
- Inner keeps only matches; left keeps every row of the left table
- How a left join shows missing data as nulls and can multiply rows
- A real example where the wrong join gave a wrong total
How would you measure whether a new checkout design is working?
A good answer covers
- One main metric (such as checkout conversion) plus guardrails like refunds or support tickets
- Comparing fairly: an A/B test, or a before-and-after with care
- How long to run it and what result would change the decision
Explain a p-value, or statistical significance, to a manager who isn't technical.
A good answer covers
- A plain-words version: how likely a difference this big is by chance alone
- What it doesn't say: how big or important the effect is
- An everyday example
You're given a messy spreadsheet of customer records. What would you check before analysing it?
A good answer covers
- Duplicates, missing values, inconsistent formats and impossible values
- Where the data came from and what each column really means
- Writing down what you cleaned, so others can repeat it
Walk me through a dashboard you built. Who used it, and what did you leave out?
A good answer covers
- Who it was for and the decision it supported
- The few metrics on it and why those
- What you removed to keep it readable, and how you knew people used it
Which tools do you use for analysis, and how do you choose between a spreadsheet, SQL and Python?
A good answer covers
- Spreadsheets for small, quick or shared work
- SQL where the data lives in a database
- Python or R for repeatable, larger or statistical work, with an example
Motivation and fit
Why do you want to work as a data analyst?
A good answer covers
- What you enjoy about the work, shown by something you've done
- The kind of business problems you want to work on
- Where you want to grow next
What do you know about our business, and which metrics would you want to look at first?
A good answer covers
- How the company makes money, in a sentence
- Two or three metrics that follow from that
- A question you'd want answered in your first month
Tips
- Practise writing SQL out loud: say what each step does before you write it.
- In a case question, ask what decision the analysis is for before you start.
- End every answer with the takeaway in one sentence, then the caveat.
- Bring one project you can talk about for ten minutes: the question, the data, the mess, the result.
Practise it out loud
A mock interview asks what data analyst interviews usually test, follows up on your answers and gives you a report. The warm-up is free, needs no account and checks five of these questions as you answer.
About this guide
Written by Orbit from what candidates and interviewers commonly describe for this kind of job. The questions are Orbit's own practice questions, not questions taken from any employer's interviews. Processes change, so check the job post and what your recruiter tells you.