Data analyst job description template: the analyst type, real SQL requirements and screening questions
On this page
"Data analyst" is one of the most stretched titles in hiring. It can mean someone who answers ad hoc business questions in SQL, someone who maintains dashboards for operations, or someone who builds data pipelines in all but name. A data analyst job description that does not say which attracts all three, and every resume lists the same tools. Below is a copy-ready template, the pay and EEO wording, how each must-have becomes a screening question and scorecard row, and the mistakes that inflate analyst requirements. The general method is in how to write a job description that screens.
Name the analyst type first
| Type | What fills the week | Must-have that separates it |
|---|---|---|
| Decision support analyst | Ad hoc questions from managers, analysis and recommendations | Has answered a vague business question with data and changed a decision |
| Reporting and BI analyst | Building and maintaining dashboards and recurring reports | Has built dashboards people use weekly, and kept them correct |
| Product or experiment analyst | Funnels, A/B tests, feature impact | Has designed or analyzed experiments and explained results honestly |
| Analytics engineer in practice | Data models, transformation code, pipeline fixes | Has built and maintained data models others query; consider titling it as such |
Also settle with the hiring manager where the data lives, who the main stakeholders are and whether the analyst will query raw tables or prepared models. Those facts tell candidates more about the job than any tool list.
The data analyst job description template
Data Analyst, [team or business area: Operations / Marketing /
Finance / Product]
[Company], [city] — [On-site / Hybrid: days / Remote within: states]
Pay: $[min]–$[max] per year, plus [bonus, if any].
[Benefits summary, if required.]
SUMMARY
You will help [operations managers / the marketing team / product
managers] make decisions with data: [answering questions such as
"why did returns rise in March?" / maintaining the weekly sales
dashboards / analyzing product experiments]. You will work in
[warehouse: e.g. a cloud data warehouse] with [BI tool] and report to
[title].
WHAT YOU WILL DO
- Turn loosely worded questions into clear analyses, and agree on the
question before you start.
- Write SQL against [raw tables / modeled data] to pull and check
the data.
- Build and maintain [dashboards / recurring reports] that [N] teams
use each week.
- Find and flag data quality problems, and work with [data
engineering / source owners] to fix them.
- Present findings in plain language with a recommendation.
- [Experiment roles: design and analyze A/B tests with product
managers.]
YOU MUST HAVE
- Written SQL in a work or substantial project setting, including
joins and aggregations across several tables.
- Answered a business question with data and explained the result to
a non-technical audience.
- Found and dealt with a data quality problem that would have changed
an answer.
- [Reporting roles: built a dashboard others relied on.]
- [Experiment roles: analyzed an experiment, including checking
whether the result was meaningful.]
NICE TO HAVE
- Experience with [BI tool], [Python or R], [dbt or similar].
- Knowledge of [domain: supply chain, subscription revenue, clinical
data].
If you meet the must-haves and none of these, please apply.
FIXED CONDITIONS
- Data access: [the role works with [customer / health / financial]
data and requires completing [privacy or security training]].
- [Background check as permitted by law, if required for data
access.]
HOW WE HIRE
[Recruiter screen; hiring manager conversation; a take-home of no
more than [N] hours with a provided dataset, or a live SQL exercise;
a walkthrough of your work with the team.]
[EEO STATEMENT]
If you need an adjustment to apply or interview, email [address].Pay range and EEO statement
Not legal advice. Pay transparency and EEO points were checked against official sources as of October 2026. Confirm the rules for each location the role can be performed from.
Pay. Leave the bracketed range until it is approved on the requisition. Remote analyst roles open in several states can trigger several pay transparency laws at once; see pay transparency laws by state. For an outside reference, be careful: the BLS occupation list has no occupation titled "data analyst". Depending on the work, the closest occupations in its OEWS program may be Operations Research Analysts (SOC 15-2031) or Data Scientists (SOC 15-2051), and neither matches a reporting analyst well. Use them as a loose check only; this page does not quote figures.
EEO statement. The EEOC lists the federal protected characteristics, including age (40 or older), and gives "recent college graduates" as an example of ad wording that may discourage older applicants (EEOC). A common closing statement:
[Company] is an equal opportunity employer. We consider qualified
applicants without regard to race, color, religion, sex (including
pregnancy, sexual orientation and gender identity), national origin,
age, disability, genetic information, protected veteran status, or
any other characteristic protected by federal, state or local law.From must-haves to screening questions and scorecard rows
A recruiter does not need to read SQL to screen these must-haves. Each question asks for a specific piece of past work, and depth shows in the detail.
| Must-have | Screening question | Scorecard competency | Evidence of a strong answer |
|---|---|---|---|
| Written SQL across several tables | "Describe the last query you wrote that joined several tables. What were you trying to find, and what went wrong the first time?" | SQL and data handling | Names the tables or entities, a join problem (duplicates, missing rows) and the fix |
| Answered a business question | "Tell me about an analysis that changed what someone decided. What was the original question?" | Problem framing and impact | Reframed a vague request and names the decision that followed |
| Handled a data quality problem | "When did the data turn out to be wrong? How did you notice, and what did you tell people?" | Data quality judgment | Caught it with a check, told stakeholders, and fixed or flagged the source |
| Built a dashboard others relied on | "Who used the dashboard, how often, and what did you change after launch?" | Reporting | Knows the users and has iterated, not just built once |
| Analyzed an experiment | "What was the last test you analyzed, and how did you decide the result was real?" | Experimentation | Mentions sample size, duration or significance in their own words |
The fuller set, including how to spot tool-name dropping, is in data analyst phone screen questions, and the interview guide for data analysts takes the same competencies into the technical and stakeholder rounds. After the hire, the 30-60-90 day plan for data analysts turns the same outcomes into first-quarter goals.
Common mistakes in data analyst job descriptions
The tool wall
"SQL, Python, R, Tableau, Power BI, Looker, Excel, dbt, Spark, Airflow" as must-haves describes no real analyst. It rewards keyword-heavy resumes and tells strong candidates the team does not know what it needs. Require SQL if it is used daily; describe the rest as the environment.
Degree requirements as a proxy for statistics
"Master's in statistics required" for a reporting role screens out good analysts from other fields. If the job needs experiment design or modeling, name it as a must-have and ask about it on the screen.
A data scientist's job at an analyst's level
Asking for machine learning, pipeline engineering and executive presentations in one analyst role either mislevels the job or hides which part matters. Pick the type and title it honestly.
No word about the data or the stakeholders
Analysts judge a role by the state of the data and who they will work with. "Work with cross-functional stakeholders" says nothing; "the operations team of 12 managers, using warehouse data modeled by a two-person data engineering team" says a lot.
An unbounded take-home
A multi-day exercise filters for free time, not skill. State the time limit in the posting and give everyone the same dataset and rubric.
Before you publish
- The analyst type, stakeholders and data environment are in the summary.
- SQL is a must-have only if used daily; other tools are described, not required.
- Data access conditions and the exercise length are stated.
- An approved pay range, the EEO statement and an accommodation contact close the posting.
If you paste the finished posting into Interview Signal, it builds the screen's question guide from the must-haves, and the scorecard quotes what the candidate actually said about their queries and data problems.
Questions people ask
Which tools should a data analyst job description require?
Require SQL if the analyst will query data directly, which most do, and name the BI tool as the environment rather than a gate. Analysts move between BI tools quickly; the harder skills are writing correct queries, checking data quality and turning a vague question into a clear answer.
Should a data analyst posting require a degree in statistics or computer science?
Usually not. Many strong analysts come from economics, social sciences, finance, operations or self-teaching. If the role involves experiment design or statistical modeling, name that work as a must-have and screen for it directly instead of using a degree as a stand-in.
How long should a data analyst take-home exercise be?
State the expected time in the posting and keep it short enough that employed candidates can do it in an evening, typically a couple of hours at most. Give everyone the same dataset and rubric, and use a live walkthrough of their work as the interview rather than a second test.
What is the difference between a data analyst and a data scientist posting?
A data analyst mostly answers business questions with existing data, through queries, dashboards and analysis. A data scientist role usually adds building predictive models or running statistical work in code. Many postings blur them; title the role by the work that fills most of the week.