AI & Data
How to Hire a Data Analyst
Data analysts turn raw data into answers. They define metrics, build reporting people trust, and dig into the questions that shape everyday decisions across product, operations, and leadership.
The difficulty is that most candidates can build a dashboard. Far fewer can pick the right question, spot when the data is misleading, and explain a finding clearly enough that someone changes what they do.
On this page11 sections
- 01What Does a Data Analyst Do?
- 02When Should You Hire a Data Analyst?
- 03Skills to Look For
- 04How to Write the Job Description
- 05How to Source Candidates
- 06How to Screen Candidates
- 07Data Analyst Interview Questions
- 08How to Evaluate Candidates
- 09Common Hiring Mistakes
- 10Hiring a Data Analyst: FAQ
- 11How Napplied Helps
The short answer
To hire a data analyst, test whether candidates can turn a messy question into a decision, not only whether they write correct SQL. Run a background call, a short SQL exercise on realistic data, an analysis case with an ambiguous business question, and a presentation of the findings to non-technical stakeholders. Weigh SQL, business thinking, rigour and communication equally.
What Does a Data Analyst Do?
A data analyst works with business teams to understand what they need to know, then queries, cleans, and analyses data to answer it. Good analysts become the trusted source of truth for how the business is performing.
Typical responsibilities
- Define and maintain core business and product metrics
- Build dashboards and recurring reports stakeholders rely on
- Run ad hoc analyses to explain changes and uncover opportunities
- Validate data quality and flag issues with tracking or pipelines
- Present findings and recommendations to non-technical teams
When Should You Hire a Data Analyst?
It is usually time to hire when one or more of these is true:
- Teams argue about numbers because nobody trusts the dashboards
- Engineers or founders spend hours each week pulling data by hand
- You are making pricing, growth, or operational decisions without clear data
- You have data in a warehouse but no one turning it into insight
Skills to Look For
Technical skills
- Strong SQL: joins, window functions, and debugging queries
- BI tools such as Looker, Tableau, Power BI, or Metabase
- Spreadsheet modelling for quick analysis and forecasting
- Metric design: definitions, funnels, cohorts, retention
- Basic statistics and awareness of common analytical pitfalls
- Python or R for analysis is a plus, not always a must
How they work
- Asks what decision an analysis will inform before starting
- Communicates findings simply, with honest caveats
- Is sceptical of surprising numbers and checks them
How to Write the Job Description
Describe the teams the analyst will support and the questions they will answer. Be clear about the balance between building reporting and doing deeper analysis.
What to include
- The teams and stakeholders the analyst will partner with
- Your data stack: warehouse, BI tools, and data engineering support
- Examples of questions they will be expected to answer
- Balance between dashboards, ad hoc work, and deeper analysis
- Compensation range and level
How to Source Candidates
Strong analysts come from finance, operations, consulting, and science as well as data teams. Look for evidence of analysis that led to a decision.
- Operations and finance teams
- Strong business context and comfort with numbers
- Analytics communities
- People sharing dashboards, SQL tips, and case studies
- Public portfolios
- Published dashboards or analyses that show reasoning
- Bootcamp graduates
- For junior roles, paired with a strong senior analyst
How to Screen Candidates
A focused process with a clear purpose for each stage:
- Background callDiscuss an analysis they ran and the decision it influenced.
- SQL exerciseRealistic tables with a few data quality traps; assess correctness and approach.
- Analysis caseExplain a sudden drop in a key metric and recommend what to investigate next.
- PresentationShare findings with a non-technical interviewer and handle follow-up questions.
Data Analyst Interview Questions
- Tests: SQL and data skills
Weekly sign-ups dropped 20% overnight. How do you investigate?
What good looks like: Checks tracking and data issues first, then segments by channel, platform, and region.
- Tests: Business thinking
How would you define an “active user” for our product?
What good looks like: Ties the definition to real value, considers edge cases, and documents it.
- Tests: Communication
Tell me about a time your analysis changed someone’s mind.
What good looks like: Clear business impact and how they communicated it.
- Tests: Business thinking
A stakeholder wants a dashboard with 40 metrics. What do you do?
What good looks like: Finds the underlying decisions and focuses on the few metrics that matter.
- Tests: Rigour
How do you make sure your numbers are right before sharing them?
What good looks like: Sanity checks, reconciling with known totals, and peer review.
How to Evaluate Candidates
Score every candidate against the same criteria, agreed before interviews start. Each criterion lists the interview questions that test it, so every score traces back to a specific answer.
| Criterion | Strong signal | Red flag |
|---|---|---|
| SQL and data skillsTested by Q1 | Accurate, readable queries; catches data issues | Correct-looking results from flawed joins |
| Business thinkingTested by Q2, Q4 | Starts from the decision, not the data | Produces numbers without a point |
| RigourTested by Q5 | Validates surprising results before sharing | Reports whatever the query returns |
| CommunicationTested by Q3 | Clear story, simple visuals, honest caveats | Cluttered charts and unclear takeaways |
Common Hiring Mistakes
- Testing only SQLSQL is the entry ticket. The real job is judgment and communication.
- Confusing analyst and data scientist rolesBe clear whether you need reporting and insight or modelling and experimentation.
- No data foundationWithout reliable tracking and a warehouse, analysts spend their time fixing pipelines.
- Treating analysts as a ticket queueEmbed them with teams so they can ask better questions.
Hiring a Data Analyst: FAQ
- What is the difference between a data analyst and a data scientist?
- A data analyst answers business questions with existing data: SQL, dashboards, metric definitions and clear recommendations. A data scientist more often builds statistical or machine learning models and runs experiments. If your questions are about what happened and why, hire an analyst. If they are about prediction or causal testing at scale, you need a data scientist.
- What SQL test should I give a data analyst?
- Use a short exercise on data that looks like yours: two or three tables to join, with some nulls, duplicates and an awkward date field. Ask a few questions that build on each other, ending with one that requires a judgment call. Watch how candidates check their results, not only whether the query runs.
- When should a startup hire its first data analyst?
- Hire when leaders regularly make decisions without trusted numbers, or when engineers spend real time pulling ad hoc reports. Make sure basic tracking and a warehouse exist first, or the analyst will spend months on data plumbing. If neither is in place, a data engineer or analytics engineer may be the better first hire.
- How long should a data analyst take-home be?
- Two to three hours at most, and pay for it if it runs longer. Provide a realistic dataset and an open question, and ask for a short written recommendation rather than a polished deck. The quality of the reasoning and honesty about the data’s limits matter more than presentation.
- How do you assess a data analyst’s communication skills?
- Ask the candidate to present their analysis case to someone outside data, then ask what a busy executive would ask: so what, how sure are you, and what should we do. Strong analysts lead with the answer, state uncertainty plainly and do not bury the point in methodology.