How to Interview a Data Analyst: A Hiring Manager’s Complete Guide

To hire a Data Analyst effectively, evaluate three distinct capabilities in sequence: technical skills (SQL, data visualisation, statistical reasoning), business judgment (translating vague questions into structured analyzes), and communication (explaining findings to non-technical stakeholders). The most reliable process runs four stages, recruiter screen, hiring manager screen, technical assessment, and interview loop. Most first-year failures are not technical failures: they are communication and judgment failures that a purely technical interview process is not designed to detect.

Here is how that failure plays out in practice. A company posts a Data Analyst role, receives 200 applications, moves six candidates through a full interview loop, extends one offer, and watches that person fail their 90-day review because nobody verified whether they could translate data into decisions a stakeholder could act on. The resume said SQL. The candidate could write SQL. But that was never really the job.

📋 At a Glance: Data Analyst Hiring

Core skills to evaluate	SQL, data visualisation, statistical reasoning, business communication
Most common hiring mistake	Testing only technical skills; neglecting business judgment and communication
Recommended interview stages	Recruiter screen → Hiring manager screen → Technical assessment → Interview loop
Key failure mode	Analyst can write queries but cannot translate findings into stakeholder decisions
Senior vs. mid-level difference	Senior analysts drive strategy and influence without authority; mid-level analysts execute independently

Who this guide is for: This guide is written for hiring managers without a data background, HR business partners and talent acquisition leads managing Data & AI roles, and founders or business leaders hiring their first or second data analyst. It does not assume any technical knowledge. If you can follow a business conversation, you can use this framework.

In This Guide:

• Why Is Hiring a Data Analyst So Difficult?, The three-skill problem that causes most first-year failures

• What Does a Data Analyst Actually Do?, Core responsibilities by seniority level and what to define before you post the role

• What Technical Skills Should a Data Analyst Have?, Non-negotiables, strong differentiators, and how to test each one

• How Should You Structure the Data Analyst Interview Process?, A four-stage process with time guidelines and what to evaluate at each stage

• What Interview Questions Should You Ask a Data Analyst?, Six specific questions with green flags, red flags, and what each one reveals

• What Are the Red Flags and Green Flags When Hiring a Data Analyst?
• How Is Hiring a Senior or Principal Data Analyst Different?
• Should I Hire a Data Analyst, a Data Scientist, or an Analytics Engineer?
• Frequently Asked Questions

Why Is Hiring a Data Analyst So Difficult?

Hiring a Data Analyst is difficult for three reasons: the role is inconsistently defined across organisations, it demands three distinct skill types simultaneously (technical ability, business judgment, and communication), and most interview processes test only one of these three areas, which is the primary driver of first-year failure.

At one company the title means building dashboards in Tableau; at another it means writing complex SQL pipelines, running A/B tests, and advising the executive team on pricing strategy. This definitional inconsistency means candidates arrive with wildly different experience profiles, and hiring managers without a data background have no reliable benchmark for what "good" looks like.

The deeper problem is structural. A strong Data Analyst needs technical depth and business judgment and communication ability. Most hiring processes test one of these three things reasonably well and neglect the other two. The result is analysts who can write beautiful queries but cannot talk to a stakeholder, or analysts who are excellent communicators but cannot get data out of a database without help. Getting all three right requires a structured process, not a gut-feel conversation.

What Does a Data Analyst Actually Do? (And What Should You Be Hiring For?)

A Data Analyst’s core function is to transform messy, incomplete, or siloed data into insights that stakeholders can use to make better decisions faster. In practice, this means writing SQL queries, building dashboards, running ad hoc analyzes, and translating vague business questions into structured, answerable problems. Before you write the job description or design the interview, define specifically which of these responsibilities this person will own in their first six months, because the weighting varies significantly by team, seniority level, and data maturity.

The day-to-day reality typically includes:

• Writing and optimising SQL queries to extract and manipulate data

• Building and maintaining dashboards and reports in tools like Tableau, Power BI, or Looker

• Conducting exploratory data analysis to surface trends, anomalies, and opportunities

• Translating vague business questions into answerable data problems

• Running ad hoc analyzes in response to business events such as a revenue drop or product launch

• Documenting data definitions and metrics to support governance

• Collaborating with data engineers on data quality and pipeline issues

The weighting of these responsibilities varies by team and seniority. A mid-level analyst should own the full analysis lifecycle independently. A senior analyst should be driving analytical strategy and mentoring others. Know which one you need before the first screening call.

Seniority Level	Core Responsibilities	What to Evaluate in the Interview
Mid-Level Analyst	Owns full analysis lifecycle independently; builds and maintains dashboards; runs ad hoc analysis	SQL depth, business question framing, stakeholder communication
Senior Analyst	Drives analytical strategy; mentors junior analysts; shapes data definitions and metrics	Strategic thinking, ability to influence without authority, quality of past business impact
Principal / Staff Analyst	Sets organisational analytical standards; bridges technical teams and executive stakeholders	Executive communication, cross-functional leadership, evidence of measurable business outcomes

What Technical Skills Should a Data Analyst Have?

A Data Analyst must demonstrate proficiency across three non-negotiable technical areas, SQL, data visualisation, and spreadsheet fluency, and ideally brings additional depth in Python, statistical reasoning, dbt, and data modelling.

Non-Negotiable for Most Roles

SQL. This is the single most universal requirement. Do not just confirm they know SQL. Test depth. Can they use window functions like RANK, ROW_NUMBER, and LAG? Do they understand CTEs, subqueries, and query optimisation? Can they apply JOINs across multiple tables using real business logic?

Data visualisation tools. Candidates should have hands-on experience with at least one major business intelligence (BI) platform.

Key BI Tools a Data Analyst Should Know (and How to Evaluate Familiarity)

Tableau is a self-service business intelligence and data visualisation platform owned by Salesforce. It is the most commonly specified BI tool in Data Analyst job descriptions globally as of 2025, and is widely deployed across financial services, retail, and healthcare. When evaluating a candidate’s Tableau experience, ask them to describe a dashboard they built for a non-technical audience and the design decisions they made, not just whether they can use the tool.

Microsoft Power BI is a cloud-based business analytics service embedded within the Microsoft 365 ecosystem. It holds the largest installed base of any BI tool by volume and is the dominant platform in organisations standardised on Microsoft infrastructure. Candidates claiming Power BI proficiency should be able to describe experience with DAX measures and data model relationships, not just drag-and-drop report building.

Looker is a web-based business intelligence platform acquired by Google in 2019, now part of Google Cloud. Its defining feature is LookML, a proprietary modelling layer that allows data teams to define metrics and business logic centrally, so that all reports pull from a single governed source of truth. Familiarity with Looker is a strong signal that a candidate has worked in a data-mature, engineering-led analytics environment.

How to evaluate BI tool depth: Do not confirm tool familiarity, confirm tool output. Ask candidates to describe the most complex dashboard they have built, who used it, and what business decision it informed. The answer tells you far more than a tool checklist.

Spreadsheet fluency. Excel and Google Sheets remain critical, particularly for collaborating with finance and operations teams. Test for pivot tables, XLOOKUP, INDEX-MATCH, SUMIFS, and light scenario modelling.

Strong Differentiators

Python or R. Not universally required, but increasingly expected at mid-senior levels. Look for experience with Pandas for data manipulation and comfort with basic statistical modelling such as regression and cohort analysis.

Statistical reasoning. Candidates do not need a PhD, but they must understand distributions, hypothesis testing, A/B test interpretation, and the difference between correlation and causation. This matters enormously in a business context.

dbt (Data Build Tool). dbt is an open-source transformation tool that allows analysts and analytics engineers to write, test, and version-control SQL-based data transformations directly within the data warehouse. It has become a standard component of modern analytics stacks at technology and data-mature companies. A Data Analyst candidate familiar with dbt has almost certainly worked in a collaborative, production-grade data environment and understands the analytics engineering layer that sits between raw data and business-facing reports. For senior roles, this is increasingly a baseline expectation rather than a differentiator.

Data modelling fundamentals. Can they read an entity-relationship diagram? Do they understand star schemas and dimensional modelling? This knowledge separates analysts who build on top of data from those who understand how it is structured.

How Should You Structure the Data Analyst Interview Process?

A reliable Data Analyst interview process runs four stages: a recruiter screen, a hiring manager screen, a technical assessment, and a focused interview loop covering technical depth, stakeholder communication, and behavioural judgment.

Stage 1: Recruiter Screen

Verify baseline fit, compensation alignment, and logistics. More importantly, listen to how they communicate. An analyst who cannot explain their work clearly to a recruiter will not explain it clearly to your CFO either.

Stage 2: Hiring Manager Screen

Explore analytical thinking, business judgment, and culture fit at a high level. This is your conversation. Ask them to walk you through their most impactful analysis. Ask how they define a good question before starting any work. Ask what their day-to-day relationship with business stakeholders looks like. Listen for specificity. Vague answers here predict vague work later.

Stage 3: Technical Assessment

Validate hands-on capability.

For the technical assessment, use a structured SQL assessment platform such as StrataScratch, an online platform providing real-world SQL and Python interview questions sourced from actual data science and data analyst interviews at technology companies, commonly used by both candidates preparing for interviews and hiring teams building structured technical assessments, or a shared SQL editor with a sanitised company dataset. Focus on problems that require real business logic, not just syntax recall.

Stage 4: Interview Loop (Two to Three Hours Total)

Run three focused conversations:

• Technical interview: SQL, statistics, and tool proficiency in depth

• Stakeholder interview: Can they explain data clearly to a non-technical person? Include someone from a business function such as marketing or finance in this one

• Behavioural interview: How do they handle ambiguity, disagreement with a stakeholder, or a data quality crisis?

How Long Does It Take to Hire a Data Analyst?

Stage	Key Risk If Rushed
Recruiter screen	Low risk, early stage
Hiring manager screen	Missing a clear signal on communication ability
Technical assessment	Unfair results if time pressure is too high
Interview loop	Candidate accepts competing offer during scheduling delays
Total	Longest failure point: internal decision-making after the loop

What Interview Questions Should You Ask a Data Analyst?

The six questions below cover the full evaluation surface for a Data Analyst interview, testing SQL depth, data quality judgment, business decomposition, statistical rigour, dashboard design, and problem definition, in that order. Each question includes a defined green flag and red flag response so that interviewers without a technical background can evaluate answers consistently.

What SQL Question Should I Ask a Data Analyst in an Interview?

Ask: "Write a query to find the top 10 customers by total spend in the last 90 days who made at least three separate purchases."

This question tests whether a candidate can apply SQL to a real business problem, not just recall syntax. A strong candidate will ask clarifying questions before writing a single line of code.

✅ Green flag: They immediately ask clarifying questions, What counts as a purchase? Is the date field a timestamp or a date?, talk through their logic out loud, and mention they would validate the output against a known benchmark.

🚩 Red flag: They jump straight to writing code without asking a single question.

How Do I Test a Data Analyst’s Approach to Data Quality?

Ask: "You pull a dataset and notice 15 percent of a key field is NULL. What do you do?"

This question tests whether a candidate treats data quality as a problem to investigate, not a nuisance to work around. Strong candidates will diagnose before deciding, and will loop in the stakeholder before delivering anything.

✅ Green flag: They ask why the NULLs exist before deciding anything, upstream pipeline issue? Optional field? Data migration gap? They mention flagging the issue to the stakeholder before delivering the analysis and documenting whatever decision they make.

🚩 Red flag: "I’d just drop the NULLs." Full stop, no further reasoning.

What Are the Red Flags and Green Flags When Hiring a Data Analyst?

Red Flags

• They describe outputs, not outcomes. A candidate who says “I built a dashboard in Tableau” without explaining what decision it supported or what changed as a result is likely task-focused rather than impact-focused. Data Analysts who cannot connect their work to a business outcome are harder to manage and harder to justify.
• They cannot explain their SQL without prompting. If you ask how they approached a complex query and they struggle to walk you through their logic in plain language, that is a warning sign. Strong analysts can narrate their reasoning to a non-technical audience. That skill is central to the role.
• They blame the data for everything. Every analyst works with messy, incomplete, or contradictory data. Candidates who respond to data quality questions with frustration or deflection, rather than explaining how they diagnosed and handled the problem, will stall when conditions are not ideal.
• They have never pushed back on a stakeholder request. If every example they give involves doing exactly what was asked, be cautious. Good analysts reframe poorly formed questions and flag when a requested metric will mislead. A candidate with no examples of constructive challenge may simply execute without thinking critically.
• Vague answers about tools they claim to know. A candidate who lists Python on their CV but cannot describe a specific script they wrote, what library they used, and why, is likely overstating their proficiency. Ask for a concrete example for every tool that matters to your stack.
• No curiosity about your data environment. Candidates who ask no questions about your data infrastructure, reporting cadence, or the problems you are trying to solve are often more interested in the job title than the actual work. Lack of curiosity in the interview usually continues on the job.

Green Flags

• They quantify their impact without being prompted. When a candidate naturally says “that analysis changed how the sales team allocated territory, which contributed to a 12% uplift in close rate,” they understand that their value is in the decision they enabled, not the chart they produced.
• They ask clarifying questions before answering case questions. Before diving into a hypothetical analysis problem, strong candidates ask about the audience, the decision at stake, and the data available. This mirrors real analytical thinking and signals they will not waste time solving the wrong problem.
• They can describe a time they found something unexpected. The best analysts do not just answer the question they were given. They notice anomalies and follow them. A candidate who can walk you through a time they spotted something nobody asked them to look for, and what they did about it, is showing you genuine analytical instinct.
• They are honest about the limits of their analysis. Strong candidates volunteer caveats. They will say things like “the sample was small so I would not over-index on this” or “the data only goes back 18 months so seasonality is uncertain.” Intellectual honesty is a strong predictor of trust within a business.
• They adapt their communication style mid-conversation. Watch whether they adjust how they explain something when you signal confusion or ask a follow-up. Analysts who can shift between a technical and executive register in real time will be far more effective with your stakeholders.

How Is Hiring a Senior or Principal Data Analyst Different?

The most important shift when hiring at the Senior or Principal level is that you are no longer hiring primarily for execution. A mid-level Data Analyst needs to be able to pull and interpret data reliably. A Senior Analyst needs to decide what questions are worth asking in the first place. That distinction sounds simple but it changes almost every part of the hiring process. Your interview questions need to probe judgment, not just technique. Ask them to describe a time they identified an analytical priority that leadership had missed, or a time they told a stakeholder that the metric they were tracking was the wrong one. Those answers will tell you far more than any SQL exercise.

At the Senior and Principal level, you should also expect a meaningful uplift in autonomy and scope. A Senior Analyst should be able to own a domain, such as customer retention, revenue forecasting, or product performance, end to end. That means defining the measurement framework, building and maintaining the core reporting, and proactively surfacing insights without being directed. Principal Analysts, where that level exists, typically set analytical standards across a team, influence how data is modelled upstream, and act as a bridge between the data engineering function and the business. When you are interviewing at this level, ask explicitly about how they have influenced work beyond their own output. If all their examples are solo contributions, they may not yet be operating at the seniority you need.

Compensation and title expectations at this level also vary significantly by industry and company size, so calibration matters before you open the role. A Principal Data Analyst at a Series B startup is often doing work that would be titled Analytics Manager or Head of Analytics at a larger enterprise. Be specific in your job description about reporting lines, whether the role has any people management responsibility, and what tools and infrastructure they will be working with. Senior candidates will filter themselves out of roles that are not a genuine step forward, and they will do it quickly if the brief is vague. A precise job description saves both sides time and signals that you understand what you are actually hiring for.

Should I Hire a Data Analyst, a Data Scientist, or an Analytics Engineer?

These three roles are frequently confused, and hiring the wrong one is an expensive mistake. They share some overlap in tools and vocabulary, but they solve different problems and operate at different points in the data workflow. The table below gives you a practical way to distinguish them before you write your job description.

Dimension	Data Analyst	Data Scientist	Analytics Engineer
Primary output	Reports, dashboards, and ad hoc analysis that inform decisions	Predictive models, statistical experiments, and algorithmic recommendations	Clean, tested, documented data models that the rest of the team builds on
Core tools	SQL, Tableau, Power BI, Excel, sometimes Python or R for analysis	Python, R, Jupyter, scikit-learn, statistical modelling frameworks	dbt, SQL, version control (Git), data warehouses such as Snowflake or BigQuery
Typical stakeholder	Business leaders, operations teams, finance, marketing	Product teams, engineering, senior leadership for strategic bets	Data Analysts and Data Scientists who consume the data they prepare
Question they answer	What happened, why did it happen, and what should we do about it?	What is likely to happen next, and how can we build a system to act on that?	Is our data accurate, consistent, and structured so others can answer their questions reliably?
When you need them	You have business questions that data should be answering but nobody is answering them systematically	You have a specific prediction or automation problem that rules-based reporting cannot solve	Your analysts are spending more time cleaning and transforming data than analysing it

If your team is spending hours in spreadsheets, stakeholders are making decisions without reliable data, and nobody owns your core business metrics, hire a Data Analyst first. If your analysts are constantly rebuilding the same data pipelines or fighting inconsistent numbers across tools, an Analytics Engineer will unblock everyone faster than adding another analyst. A Data Scientist is the right hire when you have a specific modelling problem, such as churn prediction, demand forecasting, or recommendation logic, and you already have clean, accessible data for them to work with. Hiring a Data Scientist into a data environment that is not yet reliable is one of the most common and costly sequencing mistakes in analytics team building.

Frequently Asked Questions

How long should a Data Analyst interview process take?

For most Data Analyst roles, three to four stages is appropriate. A longer process than that increases the risk of losing strong candidates to faster-moving employers. The stages should include a recruiter screen, a hiring manager conversation focused on experience and fit, a practical skills assessment, and a final interview with a senior stakeholder or team member. Avoid adding stages simply to build consensus internally.

What SQL level should I require for a Data Analyst role?

For most Data Analyst roles, you should expect confident proficiency with SELECT statements, JOINs across multiple tables, GROUP BY aggregations, subqueries, and window functions such as ROW_NUMBER and LAG. Candidates who cannot write a query involving a JOIN and a GROUP BY without heavy assistance are not ready for independent analytical work. Candidates who can optimise queries and explain execution plans are operating at a senior level.

Is a degree in a quantitative subject required to hire a strong Data Analyst?

No. While degrees in mathematics, statistics, economics, or computer science are common among strong Data Analysts, they are not a reliable filter on their own. Many highly effective analysts are self-taught or come from fields such as social science, business, or journalism. Portfolio work, practical assessments, and structured interview questions will tell you far more about capability than the subject of a degree.

How do I assess whether a Data Analyst candidate can communicate with non-technical stakeholders?

The most reliable method is to ask them to explain a past analysis to you as if you have no data background, and then observe whether they lead with the business implication or the methodology. Strong communicators start with the insight and the decision it supports. They use plain language, avoid jargon, and check for understanding. You can also ask them to describe a time a stakeholder misunderstood their analysis and what they did about it.

What is a reasonable salary range for a Data Analyst in 2025?

In the United Kingdom, Data Analyst salaries in 2025 typically range from around 30,000 pounds for entry-level roles to 65,000 pounds or more for Senior Analysts in London or high-demand sectors such as fintech, AI, and consulting. In the United States, the range runs from approximately 60,000 dollars at the junior end to over 110,000 dollars for senior talent in major markets. Compensation varies considerably by sector, company size, and whether the role is fully remote, hybrid, or on-site.

What is the difference between a Data Analyst and a Business Analyst?

A Data Analyst’s primary output is quantitative insight drawn from data, typically involving SQL, BI tools, and statistical reasoning. A Business Analyst’s work is more process and requirements focused, often bridging between business stakeholders and technology teams to define what systems or workflows should do. In practice, the roles overlap at many companies, but if your core need is to interpret data and surface insight, hire a Data Analyst. If your core need is to document requirements and manage change across systems, hire a Business Analyst.

Related interview guides

• How to Interview a Data Scientist
• How to Interview a Analytics Engineer
• How to Interview a Data Engineer
• How to Interview a Head of Data

Hiring for a Data or AI role and want a specialist partner? Explore our recruiting services or book a 15-minute call.