How to Write a Data Analyst CV
3 min read·Updated 3 Nov 2025
Data analysis is a service role, and the CVs that work say so. Nobody is hiring someone who can produce a dashboard; they are hiring someone whose dashboards get used, whose numbers survive scrutiny from finance, and who can tell an operations manager something they did not already believe.
That has a direct consequence for how bullet points end. "Built a dashboard" is incomplete. "Built the dashboard now used weekly by 40 store managers, replacing a spreadsheet that took a day to assemble" is a complete piece of evidence.
SQL: name the constructs
"Advanced SQL" appears on almost every analytics CV and is tested in almost every interview. Naming what you actually do is both more credible and more informative.
"SQL (window functions, CTEs, query tuning)" tells a reader your level without you having to claim it. It also signals that you know there is a level above joins.
Dashboards need an audience
Weak
Created dashboards in Power BI.
Stronger
Built the regional performance dashboard now used weekly by 40 store managers, replacing a manually assembled spreadsheet that took a full day of analyst time each Monday.
Names the audience, the cadence and the manual process it retired. Adoption is the hard part of reporting work, and this shows it happened.
Findings that changed something
The most persuasive analytics bullet is one where the analysis overturned an assumption. These are memorable, hard to fabricate and immediately show judgement.
Weak
Analysed customer data to identify trends.
Stronger
Segmented 1.2m customers by recency and frequency, which showed a third of "lapsed" accounts had simply changed payment method; correcting the definition raised reported active customers by 8% and changed the retention target for the year.
A finding that was surprising, quantified, and had a consequence beyond the report itself.
Data quality is most of the job and never on the CV
Cleaning, reconciling and chasing source-system problems occupy most of an analyst's week, and almost nobody writes about them. That makes it an easy way to stand out — and it demonstrates the diagnostic skill that separates an analyst from a report builder.
Describe a variance you traced to its root cause and the control you added so it stopped recurring.
Statistical honesty
If you have run experiments, say so — and if any of them led to a recommendation not to ship, say that too. "Ran the readout on 11 pricing experiments, including two where the recommendation was not to ship because the observed lift sat inside the confidence interval" demonstrates statistical discipline far more convincingly than listing hypothesis testing as a skill.
Common mistakes
- Course-project datasets. The Titanic dataset and the sample retail warehouse are instantly recognisable and add nothing.
- Listing four BI tools. It suggests four trials rather than one competence.
- Charts embedded as images with the key numbers inside them, where neither a parser nor a phone reader can get at them.
- No mention of who asked the question or what they did with the answer.
Questions
Power BI or Tableau?
Do I need Python?
How do I get experience without an analytics job?
The same structure, applied to a field
Each of these works through a complete fictional CV for that profession, including the conventions where the general advice above does not hold.
- Data AnalystAn analytics CV built on a short core of tested skills, with numbers only where the candidate actually measured them.
- Business AnalystA BA CV that shows elicitation, process modelling and requirements that survived contact with a build team.
- Financial AnalystAn FP&A CV built around forecast accuracy, the decisions your analysis supported and the models you own.
- Data ScientistA data science CV that shows deployed models, honest evaluation and the business decision the model supports.
Put this into practice
The editor keeps your CV as structured content, so the section order and wording decisions in this guide are easy to apply and easy to change again.