Resume for a Data Analyst

What employers look for

Recruiters filter on SQL first, then on the specific BI product in the posting, then on a spreadsheet term (pivot tables, lookups, modeling), and then on whether a statistics or programming line exists at all. Managers read differently. They look for the question behind each bullet: what did the business want to know, what data did you pull, what did you find, and what happened because of it. They also check for the unglamorous work, since most of an analyst's week is cleaning data, reconciling two sources that disagree, and documenting definitions so a metric means the same thing next quarter.

In the interview, expect a live SQL exercise (joins, aggregation, window functions), a take-home with a messy file, and a conversation about a time your finding was unwelcome. The resume should set those up: name the join and window-function work, mention a cleaning or reconciliation project, and include one bullet where the result changed a decision.

SQL

Say which dialect and what you did with it: multi-table joins, window functions, CTEs, query tuning on tables of a stated size. 'Wrote SQL' is the weakest possible phrasing for the most-searched keyword on the page.

Spreadsheets (Excel or Sheets)

Name the features that matter in analyst work: pivot tables, lookups, data validation, and a model you built. A bullet about a forecasting workbook a finance team relied on is worth more than 'advanced Excel'.

BI and dashboard tools

Common ones are Tableau, Power BI and Looker. Name the one the posting names, then describe a dashboard you built, who used it, how often, and what it replaced.

Statistics

Hypothesis tests, confidence intervals, regression and A/B test design are the basics managers check. Attach one to a real analysis you ran and say what the result meant for the decision.

Python or R

Optional for many analyst postings, required for some. If you have it, state the libraries (pandas and matplotlib, or the R equivalents) and a task you automated or an analysis you could not have done in SQL alone.

Data cleaning and quality

Describe a reconciliation or deduplication job with the volume involved and the error it prevented. This is the work that separates analysts who have shipped from those who have only studied.

Stakeholder communication

Reference the audience: a weekly readout to a sales director, a quarterly deck for leadership, documentation of metric definitions. Interviewers ask how you handle a stakeholder who dislikes the answer; give yourself a story.

Data modeling and ETL basics

If you built or maintained a reporting table, a scheduled job or a metrics layer, say so. It signals you understand where numbers come from, not just how to chart them.

Sample summary and bullets

A portfolio a hiring manager will open

For an analyst, a portfolio is two or three finished analyses that someone can read in ten minutes. Each should have a stated question, the data source (a public dataset is fine), the method, a chart or two, and a plain-English conclusion. Host them where a link opens without a login: a personal site, a notebook viewer, or a public dashboard page. Put the link in the header of the resume next to your email.

Pick projects that look like the job. A marketing analytics posting wants a funnel or a campaign analysis; an operations posting wants a throughput or staffing model. If your professional work is confidential, rebuild the shape of one analysis on public data and say that is what you did. Leave out course assignments that only show you can follow instructions.

• One SQL-heavy piece, with the queries visible.
• One dashboard with a short note on who it is for and what decision it supports.
• One written analysis with a recommendation, not just charts.

Common mistakes on analyst resumes

The most frequent problem is the tool list without outcomes: a line of twelve products and no sentence about what any of them produced. The second is the unexplained metric; a bullet that says 'improved reporting accuracy' with no number, baseline or audience invites a hard interview question. The third is burying SQL in a skills block at the bottom while the top of the page talks about a passion for data. Put the query work up front.

• Do not call yourself a data scientist on a data analyst application; the titles are read differently.
• Avoid claiming 'advanced Excel' without naming a feature or a workbook you built.
• Describe dashboards by audience and use, not by color or number of charts.
• Spell the BI tool exactly as the posting does; a keyword search is literal.
• State the dataset size when you mention tuning, cleaning or modeling work.

Tailor it to the posting

Analyst postings vary mostly in the stack and the domain. One wants Power BI and a finance background; the next wants Looker, Python and product metrics. Read the posting's requirements and preferred qualifications as two lists. Everything on the required list that you have should appear on page one, in the posting's words, each tied to a bullet. Preferred items you have go next. Then look at the domain words (churn, recurring revenue, conversion, inventory turns) and use the ones that are true of your work.

ClearHire generates the draft against the selected posting, drawing only on the documents you supplied, and you check and edit the result before it is used. It is a draft aimed at the posting's wording; the judgment about what is accurate stays with you.

1. List the posting's tools and domain terms, then mark which ones your experience honestly supports.
2. Open with a summary that names the domain and the top two tools in the posting, nothing else.
3. Reorder bullets so the ones matching the role's core responsibility (reporting, experimentation, forecasting) come first in your most recent job.
4. Replace generic verbs with the posting's own: if it says 'build and maintain dashboards', your bullet should say you built and maintained dashboards.
5. Add a portfolio link that points at the piece closest to the posting's domain.
6. Trim roles with no analytical content to a single line, or drop them if the page is tight.