Data analyst resume: tools in Skills, decisions in Experience

Analyst postings in Greenhouse almost always list SQL in the first screen. Workday skills matching at banks, retailers, and health systems does the same. If SQL only appears inside a dense bullet about “analyzing large datasets,” some extracts never tokenize it as a skill. Put SQL, the warehouse (Snowflake, BigQuery, Redshift, Databricks SQL), and the BI tool under a Skills heading. Then write bullets that start from a question the business asked, not from a list of joins.

A useful bullet names the metric, the grain, who consumed it, and a result you can defend: “Weekly SKU-level stockout dashboard in Tableau for 12 planners; after launch, emergency POs dropped from 9 a week to 4 (ops stand-up, Q1 2025).” That is analysis. “Used SQL, Python, and Tableau to derive insights” is a tools dump that could sit on anyone’s file. Stack belongs once in Skills and once, lightly, in the line that used it — the same split as a software engineer resume.

Match the title only when it is true. Business analyst work (requirements, process maps, UAT) is not a synonym for writing warehouse SQL. Data analyst work is not automatically “data scientist.” If the posting says Data Analyst and your badge said Business Analyst, you can write “Business Analyst (SQL reporting / Tableau)” when that was the work — you cannot invent PyTorch. Workday stores the title field; a hiring manager still reads the bullets. Tailor to the job means copying nouns you can demo, not restyling the badge.

What to copy from the posting

• Warehouses and BI: Snowflake, BigQuery, Redshift, Synapse, Tableau, Power BI, Looker, Qlik, Excel Power Query.
• Languages and transforms: SQL, Python, R, dbt, Airflow — only if you wrote or owned them.
• Domain nouns: LTV, churn, census, claims, SKU, funnel, A/B — the words their Slack already uses.
• Do not paste the whole “nice to have” library list. Humans notice. Interviews will ask you to write a window function.

Two tracks — product analytics and finance FP&A analyst — mean two files. LinkedIn can be broader. The Workday upload cannot. Keyword lists without stuffing: resume keywords and resume skills.

Workday’s apply flow parses contact, employers, dates, education, and a skills bucket. A GitHub URL in the body may land in a website field or in the blob. Recruiters rarely clone the repo from the ATS. They might click LinkedIn. Put the URL once next to LinkedIn, then prove the work in bullets: “dbt models for the orders mart, PR reviews in GitHub, daily build in Airflow.” If the only proof is the link, Greenhouse search for “dbt” can miss you.

Tableau Public and Looker looker-studio links have the same limit. They help a human after they find you. They do not replace “SQL” under Skills. Embedded dashboard screenshots in a two-column file are pictures — see two-column resume ATS. Type the dashboard name, the grain, and the audience instead.

Tools dumps in every bullet (“SQL, Python, Pandas, NumPy, Seaborn, Tableau, Excel”) crowd out the decision. Greenhouse still indexes the words, but a hiring manager stops reading. Keep the stack in Skills. Keep one or two nouns in the bullet. Taleo is worse if the dump lives in a text box. Standard headings: Work Experience, Education, Skills, Projects (optional, for bootcamp or class work with dates). ATS resume and Workday format.

Projects vs jobs

• Paid work first, reverse chronology, month–year dates Workday can sort.
• Selected projects only if they use the posting’s stack and you can demo them. Date the project.
• Coursework lists are weak. “Capstone: SQL on 50k rows, Looker studio for a mock merchandising team” is a bullet.
• Kaggle rank is optional and easy to oversell. Prefer a business question you framed.

File: selectable PDF or .docx. PDF vs Word. After export, ATS checker. Then read autofill so SQL did not vanish from Skills.

Three edits we use constantly: retail ops, marketing funnel, finance close. If you cannot name the decision, the line is still a tools dump.

No production warehouse? Write the class or intern grain honestly: “50k-row course dataset in PostgreSQL, dashboard for a mock merchandising brief.” That still parses SQL. Inventing Snowflake tenure does not. More honest metrics: resume achievements. Line craft: resume bullet points.

Three to five bullets on the most relevant seat. Older reporting jobs can collapse to two lines plus dates. A two-page list of every ad-hoc pull is how the stockout dashboard never gets read.

Ninety minutes. Job ad, last dashboard you shipped, a query you can still run or screenshot as text. Type in the Enhance CV builder if you want headings Workday already maps.

1. Circle SQL, warehouse, BI tool, Python/R, and domain nouns in the posting.
2. New file. Name, city/region, phone, email, LinkedIn, optional GitHub URL in the body — not in a header shape.
3. Headline: true title plus domain (product analyst, finance analyst) if true — not “data wizard.”
4. Jobs reverse-chronological. Employer, title, city, month–year.
5. Under the most relevant role, eight messy bullets: questions you answered, not tools you touched.
6. Delete library lists. Keep metric, grain, audience, tool, result.
7. Rewrite keepers: verb + object + warehouse or BI noun + decision or time change.
8. Skills grouped: languages, warehouses, BI, methods (A/B, forecasting) you can defend. No stars.
9. Education: degree, school, year. Bootcamp as education or a dated projects block, not a fake employer.
10. Optional projects: two max, dated, same stack as the posting.
11. Export selectable text. Confirm SQL is highlightable under Skills.
12. Paste resume + ad into the ATS checker. Upload to Workday or Greenhouse and read Skills autofill.

Summary last: title, years, domain, one dashboard or query with a result. Professional summary examples for shape. Do not paste someone else’s 40% lift.

BA vs DA: if you are applying to BA roles, lead with requirements, process, UAT, and SQL only as reporting support. If you are applying to DA roles, lead with warehouse SQL and BI. One file that claims both equally often fails both Greenhouse filters.

Desk filter. If it is not in the table: could you recreate the query or the dashboard in a 45-minute screen?

Bootcamps: date the program under Education. Put two projects with grain and tools under Projects or under a capstone job-style block with dates. Do not list the bootcamp as three employers. Student resume and resume with no experience if paid work is thin.

Business analyst postings. Requirements, workshops, UAT scripts, process maps. SQL and Tableau only if you built the report. Do not compete as a warehouse analyst with a BA-only file — or compete as a BA with a pure SQL file. Pick the requisition.

No production access. Intern and class grain is allowed when labeled. “PostgreSQL class database, 50k rows, Looker studio for a mock merchandising brief” is better than implying Snowflake ownership.

Sensitive data. Do not publish customer-level facts. Use grain and system names: “claims-level extract in SAS, HIPAA-trained.” The posting’s noun (claims, HIPAA) still belongs in text.

Career change from ops or finance. Keep true titles. Show the first SQL or dashboard with dates. Career change resume. Do not hide the old job’s dates in a functional cloud.

When two dashboards exist, keep the one you can rebuild in a 45-minute screen: grain, warehouse, audience, and the decision that followed. A SKU-week stockout board in Tableau beats “generated insights for stakeholders.” If you cannot name the stand-up that used it, the line is still a tools dump. SQL stays under Skills so Workday tokenizes it; the bullet carries the question.

GitHub and Tableau Public help after someone finds you. They do not replace dates on paid work. Put the URL once in the body, then describe the mart, the PR, or the notebook in a dated bullet. Greenhouse search for dbt will miss a link-only file. A private repo you cannot demo should not be the only proof.

Bootcamp and class grain is allowed when labeled. “50k-row PostgreSQL set, Looker studio for a mock merchandising brief” parses as SQL. Implying you owned the revenue pipeline in Snowflake does not survive the first whiteboard. Keep two files if you are applying to BA and DA requisitions in the same week.

What to do today: pick one posting, rewrite three bullets with grain and a decision, score, upload, read Skills. Enhance CV back to builder and checker.