Data Analyst Interview Scorecard

A 6-factor weighted scorecard for hiring data analysts on analytical reasoning, SQL and data wrangling, statistical rigor, visualization and storytelling, business acumen, and stakeholder communication. Built for teams screening analysts who turn messy data into decisions, not just dashboards.

When to use this scorecard

Use this AI scorecard when you're hiring a data analyst and need someone who frames the right question and explains the answer, not just a report-builder who writes queries on request.

Use this for data analyst roles that pull, clean, and interpret data to answer business questions and inform decisions. It is the right rubric when you need someone who can frame the right question, write the query, and explain the result to a non-technical stakeholder, rather than a pure report-builder or a research scientist.

This scorecard works best on video answers paired with a take-home or live data exercise. Use the video round to probe how a candidate reasons about an ambiguous question and communicates a finding; use the exercise to confirm SQL and rigor. For heavy modeling or ML roles, raise statistical rigor or use a data scientist rubric instead.

The full scorecard

The scorecard has six weighted factors that sum to 100%: Analytical Reasoning (25%), SQL & Data Wrangling (20%), Statistical Rigor (15%), Data Visualization & Storytelling (20%), Business Acumen (10%), and Stakeholder Communication (10%).

6 factors · 100% weightage · 1–5 scoring rubric

Frames the right question, structures an approach, and reasons from data to a defensible conclusion.

What to look for

Clarifies the real business question before touching data
Breaks an ambiguous problem into a structured approach
Forms and tests hypotheses rather than fishing for patterns
Reaches conclusions the data actually supports

Score	Rating	Description
1	Poor	Jumps to a query or chart without understanding the question.
2	Needs Improvement	Answers the literal request but misses the underlying business question.
3	Satisfactory	Structures a sound approach for clear, well-defined problems.
4	Very Good	Frames ambiguous problems well and reasons carefully to a defensible answer.
5	Excellent	Reframes the question to what actually matters and reasons rigorously under ambiguity.

Writes correct, efficient queries and cleans real, messy data into something trustworthy.

What to look for

Comfortable with joins, aggregations, window functions, and CTEs
Handles nulls, duplicates, and dirty data deliberately
Validates query output instead of trusting the first result
Writes readable, maintainable queries

Score	Rating	Description
1	Poor	Struggles with basic joins and aggregations; can't clean messy data.
2	Needs Improvement	Writes simple queries but errors on multi-table logic or data quality.
3	Satisfactory	Solid everyday SQL; handles common cleaning tasks correctly.
4	Very Good	Strong, efficient SQL and a deliberate approach to messy data and validation.
5	Excellent	Fluent and efficient; anticipates data quality traps and validates rigorously.

Applies sound statistical thinking and avoids the common traps that produce confidently wrong answers.

What to look for

Knows when a difference is signal versus noise
Avoids confusing correlation with causation
Understands sample size, bias, and significance at a working level
Quantifies uncertainty instead of stating false precision

Score	Rating	Description
1	Poor	No statistical instinct; reports random noise as findings.
2	Needs Improvement	Basic averages and counts but falls for common statistical traps.
3	Satisfactory	Sound working stats; avoids the obvious traps on routine analysis.
4	Very Good	Reasons carefully about significance, bias, and causation.
5	Excellent	Strong statistical judgment; proactively flags uncertainty and confounders.

Turns analysis into a clear, honest visual narrative that drives a decision.

What to look for

Chooses the right chart for the question, not the flashiest
Builds clean, non-misleading visuals
Leads with the insight and the 'so what', not the methodology
Tailors the story to the audience's altitude

Score	Rating	Description
1	Poor	Dumps tables or cluttered charts with no clear takeaway.
2	Needs Improvement	Makes charts but they bury the insight or mislead.
3	Satisfactory	Clear, correct visuals that communicate the basic finding.
4	Very Good	Chooses the right visuals and leads with the insight and its implication.
5	Excellent	Tells a tight, honest data story that makes the decision obvious to any audience.

Connects analysis to the metrics and decisions the business actually cares about.

What to look for

Understands how the business makes money and what drives it
Prioritizes analysis by decision impact, not curiosity
Translates a finding into a recommended action
Knows which metrics matter to which stakeholder

Score	Rating	Description
1	Poor	Analyzes in a vacuum; no sense of business impact or priorities.
2	Needs Improvement	Understands the metric but not why it matters to the business.
3	Satisfactory	Connects analysis to business goals on familiar problems.
4	Very Good	Prioritizes by impact and turns findings into clear recommendations.
5	Excellent	Thinks like an owner; anticipates the decision and analyzes to inform it.

Explains technical work to non-technical stakeholders and manages analysis requests well.

What to look for

Explains methods and caveats without jargon
Pushes back on a vague request to get a better question
Sets realistic expectations on scope and timing
Documents assumptions so results are reproducible

Score	Rating	Description
1	Poor	Drowns stakeholders in jargon or can't explain what they did.
2	Needs Improvement	Communicates results but not caveats, assumptions, or limits.
3	Satisfactory	Explains findings clearly to a technical-friendly audience.
4	Very Good	Translates cleanly for non-technical stakeholders and refines vague requests.
5	Excellent	Trusted partner to the business; turns fuzzy asks into sharp, well-scoped analysis.

Sample interview questions linked to factors

Use these five scenario questions to probe all six factors. Each maps to the factors it most directly surfaces, so panel scoring stays consistent across technical and non-technical reviewers.

Question	Factors evaluated
A stakeholder says 'our numbers are down, can you look into it?' What do you do before writing a single query?	Analytical Reasoning · Stakeholder Communication
Walk me through how you'd find out why a key metric dropped last month, from raw data to recommendation.	Analytical Reasoning · SQL & Data Wrangling · Business Acumen
You run an analysis and the result looks surprising. How do you decide whether it's real before sharing it?	Statistical Rigor · SQL & Data Wrangling
Tell me about an analysis you presented to non-technical leaders. How did you structure it?	Data Visualization & Storytelling · Stakeholder Communication
Describe a time your analysis changed a decision. What was the finding and how did you make the case?	Business Acumen · Data Visualization & Storytelling

Customization notes

Adjust weightages by analyst type. Product analysts weight statistics higher; BI analysts weight SQL and visualization higher; embedded analysts weight communication and business acumen higher.

Product / growth analyst
Raise Statistical Rigor to 20% and Business Acumen to 15%, reducing Stakeholder Communication to 5%. Experimentation and A/B testing put more weight on statistical judgment and product sense.
BI / reporting analyst
Raise SQL & Data Wrangling and Data Visualization to 25% each and reduce Statistical Rigor to 5%. Dashboard-heavy roles reward query depth and clean, scalable visuals over inferential statistics.
Analyst embedded with business teams
Raise Business Acumen and Stakeholder Communication to 15% and 20% and reduce SQL to 15%. When the analyst sits with marketing, finance, or ops, framing and communication outweigh raw query depth.
Junior / entry-level analyst
Raise Analytical Reasoning to 30% and reduce Business Acumen to 5%. Junior hires are bets on reasoning and learning speed; business context is teachable on the job.

Why a weighted rubric matters for data analysts

Why analytical reasoning and data storytelling carry 45% of the score, and what structured screening changes for a role where the wrong answer is delivered confidently.

The expensive failure mode for a data analyst is being confidently wrong, or being technically correct but unable to make anyone act. Weighting Analytical Reasoning and Data Visualization & Storytelling at 45% combined targets the bookends that matter most: framing the right question and landing the answer. SQL is necessary but the most over-indexed factor in analyst hiring, since query skill is testable and teachable while reasoning is neither.

Lower variance

Adding a scoring rubric to interviews materially reduces interviewer rating variance and bias compared with unstructured judgment calls.

Highhouse, 2008 (Industrial and Organizational Psychology)

#1 priority

Quality of hire is the top hiring priority for talent leaders, and structured interviews are the method most cited for improving it.

LinkedIn Future of Recruiting Report, 2024

Up to 30%

Bad hires cost employers up to 30% of the employee's first-year earnings, which is why structured screening pays back fast.

U.S. Department of Labor (via SHRM)

Frequently asked questions about hiring data analysts

Common questions when using this AI scorecard to hire data analysts, from balancing SQL skill with reasoning to adapting it for product versus BI roles.

Should I weight SQL higher since it's the most testable skill?

It's tempting because SQL is easy to test, but resist it. Query skill is teachable and a take-home exercise verifies it cheaply. The factors that actually predict a strong analyst, framing the question and communicating the answer, are harder to test and harder to teach, which is why this rubric weights reasoning and storytelling above SQL.

How is a data analyst scorecard different from a data scientist one?

This rubric weights analytical reasoning, SQL, and communication for descriptive and diagnostic work. A data scientist role would weight statistical rigor and add machine learning and modeling factors. If the role involves building predictive models in production, raise Statistical Rigor to 25% and add an ML factor, or use a dedicated data scientist rubric.

What's the clearest red flag across these factors?

High SQL & Data Wrangling paired with low Analytical Reasoning. It's the classic query-writer who answers exactly what was asked, even when it's the wrong question, and never reframes toward what the business actually needs to know. They look productive while producing analysis nobody can act on.

Can I use this to hire a junior analyst with no work history?

Yes. For junior hires, raise Analytical Reasoning to 30% and lean on the scenario questions rather than past-impact questions. You're betting on how they think and how fast they learn, since SQL fluency and business context both develop quickly on the job with good mentorship.

Related scorecards

Pair this rubric with the Software Engineer or QA Engineer scorecards for technically adjacent roles, or the SEO Specialist scorecard for analytics-heavy marketing roles.

Software Engineer Scorecard QA / Test Engineer Scorecard SEO Specialist Scorecard

Drop this scorecard into Hirevire

Use this rubric and the linked sample questions to score every video answer automatically. Hirevire's AI does the first pass, so you focus on the candidates worth your time.

See how AI Scorecards work