Skip to content

Data analyst portfolio: a project review checklist

Build a reproducible analysis with clear metric definitions, SQL checks, a useful chart and a defensible recommendation.

A data analyst portfolio should show how you moved from a question to a defensible decision. A dashboard screenshot can show presentation skill, but a reviewer also needs the definition, query, checks and limits behind it.

Choose a question with a decision

For a suggested project, ask which customer segment needs further investigation after a decline in completed orders. Define the population, completion event, date range and comparison. Explain what action the result could inform and what the dataset cannot establish.

Document and check the source

Use permitted public or synthetic data. Include a data dictionary, keys, missing-value treatment and source date. Check whether joins increase row counts unexpectedly and whether totals reconcile. Record exclusions such as test transactions so another person can understand the population.

Make the analysis reproducible

Save the SQL or spreadsheet steps and provide setup instructions. Use relative paths and remove secrets. A reviewer should be able to rerun the analysis from the supplied data and see how the final values were produced. Explain any manual step instead of hiding it.

Choose a chart that answers the question

Label units, dates and denominators. Separate an overall trend from differences between segments. Provide a short explanation and relevant limitation beside the chart. Do not claim causality from a before-and-after picture without a suitable study design.

Show how you review assisted work

If an approved AI tool helped draft SQL or text, explain how you checked its output. Test a query against known cases and inspect joins, filters and definitions. Do not upload private data to an unapproved service or claim independent work you cannot explain.

Package the evidence for hiring

Open with the question, method, result and limitation. Link the query, reproducible files and a readable visual. Keep measured outcomes only when the project actually observed them. A hypothetical recommendation should not become a fabricated revenue improvement in your CV.

Sources and next steps

Sources checked on October 2, 2026. Follow the linked documentation for current requirements.

PostgreSQL tutorial

Microsoft Power BI documentation

Entry-level analyst CV example

Frequently asked questions

Why Dataaxy

Everything you need for a focused Data & AI job search

  • Specialist job catalogue

    Explore roles across Data Analytics, Data Engineering, Data Science, Machine Learning, and AI.

  • Focused search tools

    Save a search and use job alerts to follow the roles, locations, and work setups that fit.

  • Public talent profiles

    Show recruiters your skills, availability, preferred roles, and work preferences in one place.

© Dataaxy. All rights reserved.Job data is gathered from publicly available sources or contributed by users.