About Analyst interviews
Interviews for a generalist Analyst role typically run across three to four stages and are designed to test whether you can turn messy, ambiguous business questions into defensible, data-backed recommendations. Expect a recruiter screen first, focused on motivation, salary expectations, and a sanity-check on stated tool experience (Excel, SQL, often Tableau or Power BI). The hiring manager round digs into how you actually work: how you scope a request from a stakeholder, how you handle conflicting priorities, and how you communicate findings to non-technical audiences. Most organisations include a technical or case component — frequently a take-home dataset, a live SQL test, or a slide-based case where you present an analysis and field challenge questions. A final round often covers stakeholder management and culture fit, sometimes with a senior business leader. Candidates most often stumble in two places. First, they over-index on tool mechanics and under-deliver on the 'so what' — they describe what they queried but not what decision it drove. Second, they fail the communication test: they bury the headline, can't defend an assumption, or freeze when an interviewer pushes back on their methodology. Strong candidates show structured thinking, comfort with ambiguity, numeracy under light pressure, and the instinct to quantify business impact rather than just describe activity.
Typical stages
- Recruiter screen
- Hiring manager interview
- Technical / case study (SQL test or take-home)
- Final / stakeholder & values round
Common formats
- Behavioral STAR
- Live SQL or Excel test
- Take-home data case
- Case study presentation
- Stakeholder roleplay
What hiring managers screen for
- Structured problem-solving and ability to scope an ambiguous question before diving into data
- Numeracy and data fluency — clean SQL/Excel logic plus sensible sanity-checking of results
- Clear communication that leads with the headline and translates analysis into business decisions
- Stakeholder management: handling pushback, conflicting requests, and unrealistic deadlines
- Commercial awareness — understanding why a metric matters, not just how to calculate it
Red flags to avoid
- Describing tools and queries without ever stating the decision or impact the analysis drove
- Accepting a vague request at face value with no clarifying questions or assumption-stating
- Inability to defend a methodology or sanity-check a result when challenged
- Burying the answer — presenting raw output instead of a clear recommendation
- Overclaiming on tools (e.g. listing SQL as expert) then failing basic joins or aggregation in the live test
Primary questions (14)
Behavioural
Tell me about a time your analysis changed a business decision.
Why this comes up: Hiring managers want proof your work drives outcomes, not just produces reports.
Prep pointers
- Pick an example where the decision genuinely shifted because of your finding, not one that merely confirmed existing plans.
- STAR: Situation = the business question and who owned the decision; Task = your analytical mandate; Action = method and how you communicated it; Result = the decision changed and the quantified impact.
- Lead with the business outcome, then back into the analysis — avoid opening with the SQL.
- Avoid examples where you can't articulate what would have happened without your work.
Behavioural
Describe a time you had to deliver an analysis under a tight deadline with incomplete data.
Why this comes up: Analysts routinely work with imperfect inputs and pressure, so interviewers probe how you cope.
Prep pointers
- Show how you triaged: what you delivered now versus what you flagged as caveats or follow-ups.
- STAR: Action should make explicit the trade-offs you made and how you communicated confidence levels.
- Demonstrate you stated assumptions transparently rather than presenting shaky numbers as certain.
- Avoid framing this as 'I worked late and powered through' — the skill is prioritisation, not heroics.
Behavioural
Tell me about a time a stakeholder disagreed with your findings.
Why this comes up: Analysts must defend conclusions diplomatically while staying open to being wrong.
Prep pointers
- Distinguish between disagreement on method versus disagreement on the implication for their team.
- STAR: Action should show you separated emotion from evidence and either held firm with data or genuinely revised your view.
- Highlight how you preserved the working relationship, not just won the argument.
- Avoid stories where you 'proved them wrong' with no nuance — show judgement.
Behavioural
Give an example of a recurring report or process you improved or automated.
Why this comes up: Efficiency and a bias toward removing manual toil are core analyst signals.
Prep pointers
- Quantify the time saved or error reduction, and who benefited.
- STAR: Action should describe the specific change (a query, a template, a pipeline) and how you got buy-in to adopt it.
- Mention how you ensured the improved process was sustainable after you handed it off.
- Avoid trivial examples; pick one with measurable ongoing payoff.
Technical
Walk me through how you'd write a SQL query to find the top three products by revenue per region last quarter.
Why this comes up: It tests joins, aggregation, filtering, and window functions in one realistic prompt.
Prep pointers
- Think aloud about the tables and grain you'd need before writing any SQL.
- Be ready to use a window function (ROW_NUMBER or RANK partitioned by region) and explain ties.
- State how you'd handle the date filter and any currency or refund adjustments.
- Verbalise a sanity check: do the regional totals roughly reconcile to a known number?
Technical
How do you investigate and explain a sudden 20% drop in a key metric?
Why this comes up: Root-cause / diagnostic analysis is a daily analyst task and tests structured thinking.
Prep pointers
- Lay out a structured decomposition: is it real or a data issue? Then segment by dimension.
- Mention checking for tracking/instrumentation breaks before assuming a business cause.
- Describe how you'd isolate the segment driving the drop rather than reporting the aggregate.
- Avoid jumping straight to a hypothesis — show the funnel of elimination.
Technical
How do you decide which chart or visualisation best communicates a given finding?
Why this comes up: Visualisation choice signals whether you think about the audience, not just the data.
Prep pointers
- Tie chart choice to the question being answered (trend, comparison, composition, distribution).
- Mention reducing cognitive load: clear labels, removing chartjunk, highlighting the takeaway.
- Give an example of a chart type you deliberately avoided and why.
- Avoid listing chart types generically — connect each to audience and intent.
Situational
A senior leader asks for a number by end of day, but you suspect the underlying data is unreliable. What do you do?
Why this comes up: Tests judgement, integrity, and how you balance speed against accuracy.
Prep pointers
- Show you'd deliver a figure with an explicit confidence caveat rather than silently delaying or sending bad numbers.
- Describe quick validation steps you could run within the time available.
- Frame how you'd communicate the risk upward without sounding obstructive.
- Avoid the extremes of 'just give them the number' or 'refuse until it's perfect'.
Situational
Two stakeholders give you conflicting requests with the same deadline. How do you handle it?
Why this comes up: Analysts sit between teams and must prioritise without a clear chain of command.
Prep pointers
- Explain how you'd surface the conflict to make priority a decision, not a guess.
- Reference criteria you'd use: business impact, deadline immovability, effort.
- Show you'd manage expectations on both sides rather than quietly dropping one.
- Avoid implying you'd just do both by working longer.
Situational
You're handed a vague request: 'Can you look into why sales are down?' How do you scope it?
Why this comes up: Scoping ambiguous requests is the single most common real analyst task.
Prep pointers
- Demonstrate the clarifying questions you'd ask: which segment, timeframe, comparison baseline, decision it informs.
- Show you'd agree a deliverable and timeline before doing analysis.
- Mention defining success — what answer would actually be actionable for them.
- Avoid diving into data before you've reframed the question.
Competency
How do you ensure the accuracy of your analysis before sharing it?
Why this comes up: Data quality and self-checking discipline are non-negotiable for analyst trust.
Prep pointers
- Describe concrete checks: reconciling totals, spot-checking rows, comparing against a known benchmark.
- Mention peer review or documenting assumptions as part of your process.
- Give an example of a time a check caught an error before it went out.
- Avoid vague claims like 'I just double-check' — name your actual steps.
Competency
How do you explain a complex analytical finding to a non-technical audience?
Why this comes up: Translation skill separates a strong analyst from a competent query-writer.
Prep pointers
- Show you lead with the conclusion and implication, then offer detail on demand.
- Mention tailoring language and removing jargon based on the audience.
- Give an example where simplifying actually changed how the message landed.
- Avoid suggesting you'd dumb it down — the skill is clarity, not condescension.
Competency
How do you prioritise your analytical workload when everything feels urgent?
Why this comes up: Analysts juggle many ad hoc requests and need a defensible prioritisation system.
Prep pointers
- Describe a framework (impact vs effort, alignment to OKRs) you actually use.
- Mention how you make trade-offs visible to requesters rather than absorbing everything.
- Reference distinguishing genuine urgency from perceived urgency.
- Avoid implying you simply work through a never-ending queue.
Culture fit
What kind of business questions do you find most motivating to work on?
Why this comes up: Tests genuine curiosity and whether your interests align with the team's domain.
Prep pointers
- Connect your answer to the type of problems this specific team or company faces.
- Show intellectual curiosity beyond just 'I like working with data'.
- Be honest about what energises you — manufactured enthusiasm reads as hollow.
- Avoid generic answers that would fit any analyst role anywhere.
More practice questions (15)
Technical
What's the difference between an INNER JOIN and a LEFT JOIN, and when would a LEFT JOIN cause double-counting?
Why this comes up: Tests whether your SQL fluency goes beyond syntax to understanding row multiplication.
Technical
How would you identify and handle duplicate records in a dataset?
Why this comes up: Data cleaning is a constant analyst task and duplicates silently corrupt results.
Technical
Explain the difference between correlation and causation with an example from your work.
Why this comes up: Guards against overclaiming conclusions, a frequent analyst pitfall.
Technical
When would you use a median rather than a mean to summarise data?
Why this comes up: Checks basic statistical judgement and awareness of skewed distributions.
Technical
How would you structure an Excel model to be auditable and reusable by someone else?
Why this comes up: Many analyst roles still live in Excel and reproducibility matters.
Situational
Your dashboard shows a number a stakeholder insists is wrong. How do you investigate?
Why this comes up: Tests calm diagnostic process and willingness to question your own output.
Situational
You realise after sending a report that one figure was incorrect. What do you do?
Why this comes up: Probes integrity and ownership under uncomfortable circumstances.
Situational
A stakeholder wants a metric tracked that you think is misleading. How do you respond?
Why this comes up: Tests whether you can push back constructively on flawed measurement.
Behavioural
Tell me about a time you had to learn a new tool or technique quickly for a project.
Why this comes up: Analyst tooling evolves fast and self-directed learning is essential.
Behavioural
Describe a time you found an insight nobody had asked for.
Why this comes up: Signals proactive curiosity beyond fulfilling assigned requests.
Behavioural
Tell me about a piece of analysis you're particularly proud of and why.
Why this comes up: Reveals what you value in your work and your standard for quality.
Competency
How do you keep your analytical work documented and reproducible?
Why this comes up: Reproducibility distinguishes reliable analysts from one-off contributors.
Competency
How do you stay current with data tools and analytical methods?
Why this comes up: Indicates whether you'll keep adding value as the toolset shifts.
Culture fit
How do you prefer to receive feedback on your analysis?
Why this comes up: Tests coachability and how you'll fit into a collaborative review culture.
Culture fit
What would make a data or analytics team a great place for you to work?
Why this comes up: Surfaces whether your working-style expectations match the team's reality.
Researching the Analyst role?
See the full skills, salary and market breakdown — what employers
actually want, and the biggest skills gaps.
Analyst skills & salary →
Get a prep pack tailored to your experience
describe.me matches these questions against your real work history,
flags your prep priorities, and gives you a STAR scaffold per question.
Start free →