The bar for a data analyst resume is evidence that your analysis changed a decision. SQL proficiency is assumed; impact is what differentiates.
Keywords an ATS holds for data analyst roles
These are the terms that recur across data analyst postings and appear in the search filters recruiters run. They are not a checklist to paste in — use the ones that describe work you have genuinely done, phrased the way the market phrases it.
- data analyst
- data analysis
- sql
- data visualization
- dashboard
- business intelligence
- reporting
- etl
- data modeling
- statistical analysis
- a/b testing
- forecasting
- kpi
- stakeholder management
- data quality
- cohort analysis
- segmentation
- python
- excel
Core skills to list explicitly
Skills sections are parsed into structured fields and searched directly, so a skill only mentioned inside a bullet is weaker than one listed here as well. Split them the way hiring managers think about them.
- Hard skills — SQL, Python (pandas, numpy), R, Statistical analysis, Data modelling, ETL / ELT, Dashboard design, A/B test design and analysis, Forecasting, Cohort analysis, Data quality management, Excel modelling
- Tools and platforms — PostgreSQL, MySQL, BigQuery, Snowflake, Redshift, dbt, Airflow, Tableau, Power BI, Looker, Looker Studio, Excel, Jupyter, Git, Amplitude, Mixpanel
- Working skills — Stakeholder communication, Requirements gathering, Data storytelling, Presenting to non-technical audiences, Prioritisation
The numbers that move a data analyst hiring manager
Quantified bullets consistently outperform unquantified ones, but the useful numbers are role-specific. These are the ones that get read in this field. Use only figures you can defend in an interview — the builder will leave an [ADD METRIC] placeholder rather than invent one for you.
- Rows, tables or data volume you worked across
- Number of stakeholders or teams served by your reporting
- Decision or financial outcome your analysis drove
- Time saved by automating a manual report
- Dashboard adoption — weekly active users
- Forecast accuracy or model error improvement
- Data quality issues identified and resolved
Example bullets (illustrative)
These are examples of shape and specificity, not text to copy. They belong to nobody, and the AI writers will never insert them into your resume — everything they produce comes from your own profile.
- Built the churn cohort model that identified a 3-week onboarding drop-off, informing a product change that reduced 90-day churn from 18% to 13%.
- Replaced 11 manual Excel reports with a single dbt-modelled Looker dashboard used weekly by 40 people across sales and finance, saving roughly 20 analyst-hours per month.
- Designed and analysed 14 A/B tests on the signup funnel, of which 5 shipped, producing a combined 9% lift in activation.
- Rebuilt the revenue reporting pipeline in BigQuery and dbt, cutting refresh time from 6 hours to 25 minutes and resolving a long-standing 2% discrepancy with finance.
- Partnered with the operations team to define 9 KPIs and their SQL definitions, ending a recurring disagreement about which numbers were correct.
Mistakes that recur on data analyst resumes
- Describing tools, not questions — 'Used SQL and Tableau daily' says nothing. Say what question you answered and what changed as a result.
- No stakeholder context — Analysis that nobody acted on is invisible. Name who used your work and what they did differently because of it.
- Omitting SQL from the skills section — It feels too obvious to list. It is also the single most-filtered keyword in analytics hiring, so it must appear verbatim.
- Confusing analyst and data scientist positioning — Listing deep learning on an analyst resume without production ML experience raises doubt rather than range. Position for the role you are applying to.
- Screenshots of dashboards — Images extract as nothing. Describe the dashboard in text and link to a portfolio if you have one.
Section order and structure
A technical skills block near the top is essential — SQL, Python, and your BI tool are hard filters. Keep a Projects section if your work history is short; a well-documented analysis project substitutes credibly for a first job.
Recommended template: Circuit or Atlas. Analytics hiring is portal-heavy, so stay single column.
- Contact
- Professional Summary
- Technical Skills
- Work Experience
- Projects
- Education
- Certifications
Writing the summary
State your domain, your stack and one decision you influenced. 'Data analyst, 4 years in B2B SaaS; SQL, dbt and Looker; built the churn model behind a 5-point retention improvement.'
Certifications worth listing
Certifications are frequently used as hard filters, so list every relevant one you hold with the awarding body and year. Never list one you are 'planning to take' — in progress with an exam date is acceptable and honest.
- Google Data Analytics Professional Certificate
- Microsoft Certified: Power BI Data Analyst Associate
- Tableau Desktop Specialist
- dbt Analytics Engineering Certification
- SnowPro Core
- AWS Certified Data Analytics
Frequently asked questions
- Do I need a portfolio as a data analyst?
- It helps significantly when your work history is short or your professional work is confidential. Two or three well-documented analyses with a stated question, method and conclusion beat ten notebooks.
- SQL or Python — which matters more on the resume?
- SQL, by a wide margin, for analyst roles. Python is a strong differentiator but rarely a hard requirement below senior level. List both if you have both.
- How do I move from analyst to data scientist?
- Lead with the statistical and experimental work you already do — A/B test design, forecasting, modelling — and be specific about methods. Reframing existing work honestly is more effective than adding ML keywords you cannot defend.
- Should I include Excel?
- Yes. It is still used everywhere, it is still filtered on, and omitting it to look advanced only costs you matches.
- How do I handle confidential company data in examples?
- Use relative figures and percentages rather than absolutes, and describe the mechanism. 'Reduced churn by 5 points' is publishable where the underlying revenue figure is not.