How to Build a Data Analyst Portfolio That Actually Gets UK Interviews

Most beginner Data Analyst portfolios in the UK look the same: a half-finished Titanic notebook, a Netflix dashboard, and a README written like an apology. UK recruiters open dozens of these every week, and they almost never lead to an interview.

A portfolio is not a skills checklist. It is a hiring shortcut. Its single job is to convince a busy hiring manager, in under three minutes, that you can do the work without supervision.

If you have not started applying yet, pair this guide with our practical first Data Analyst job in the UK walkthrough.

Why Most Beginner Portfolios Get Ignored

Before talking about what to build, it helps to understand why most portfolios get scrolled past in seconds. Recurring weaknesses include:

• no clear business question - just a dataset and some charts
• no UK context - data and currencies that feel disconnected from the role you want
• copy-paste tutorials with no original analysis or commentary
• no SQL anywhere - even though SQL is in nearly every UK Data Analyst job description
• broken links, blank READMEs, or dashboards that need a login
• too many shallow projects instead of three deep ones

If a hiring manager has to think too hard, they move on. Your portfolio has to make their decision easy.

The 4 Projects UK Recruiters Expect to See

You do not need ten projects. You need four projects that together prove you can clean data, query data, visualise data, and explain a decision in business language.

1. A SQL-First Analysis Project

Pick a dataset large enough to need joins (orders, customers, products is a classic). Load it into PostgreSQL or BigQuery and answer five business questions purely with SQL: things like monthly revenue trend, top customers by margin, and a cohort retention view. Commit the .sql files. This is the project recruiters open first.

If your SQL is not yet at that level, use our SQL beginner guide to close the gap.

2. A BI Dashboard with a Story

Build one dashboard in Power BI or Tableau Public. Anchor it to a single business question (for example, "why did Q3 sales drop in the South region?") and design every chart around answering it. End with a short executive summary written for a non-technical reader.

Not sure which tool to start with? Read our breakdown of Power BI vs Tableau in the UK.

3. An End-to-End Python or Excel Analysis

Take messy raw data and walk through cleaning, exploration, hypothesis, and conclusion. Use Python with pandas, or a documented Excel workbook for non-technical roles. The deliverable is a one-page write-up with three charts, three findings, and one recommendation.

4. One UK-Domain Project

This is what separates strong portfolios from generic ones. Pick a domain you can credibly speak about - retail, fintech, NHS open data, transport, energy - and build something that mirrors a real internal question. UK hiring managers respond to domain familiarity faster than to technical complexity.

How to Structure the GitHub Repo

Each project should live in its own repository. Inside, keep the structure boring and predictable:

• README.md at the root - the only file most reviewers will read
• /data - the raw and cleaned data, or a link if it is too large
• /sql or /notebooks - the working files
• /output - charts, dashboards screenshots, or a one-page PDF summary

Do not commit credentials, large binary files, or unfinished branches. Do commit a clear list of tools used and a short "how to reproduce" section.

The README Rules That Convert Into Interviews

A good README answers six questions in this order:

1. What business question does this project answer?
2. What data did you use, and where did it come from?
3. What tools and skills did you apply?
4. What did you find? (Three bullet points are enough.)
5. What would you do next with more time or more data?
6. How can someone reproduce your work?

Write it in plain English. If a non-technical friend cannot follow it, a stakeholder will not either - and that is exactly the skill the role is hiring for.

Where to Host the Portfolio

Three layers work well together for UK applications:

• GitHub for the code, SQL, and notebooks
• Tableau Public or Power BI Service for an interactive dashboard link
• A small personal site (Notion, Carrd, or a simple Next.js page) to summarise everything in one place

Put a single link to that personal site at the top of your CV and your LinkedIn About section. That is what most recruiters click first.

Common Mistakes to Avoid

• using Kaggle datasets without reframing them around a real business question
• showing 12 mediocre projects instead of 4 strong ones
• using AI to generate analysis without understanding it - interviewers will catch this in minutes
• writing READMEs that explain the code but not the decision
• forgetting to link the portfolio on LinkedIn and your CV

A 6-Week Plan to Build the Portfolio From Scratch

If you are starting today, a realistic schedule looks like this:

• Weeks 1-2: SQL-first project, polished and pushed to GitHub
• Weeks 3-4: Power BI or Tableau dashboard with a written summary
• Week 5: End-to-end Python or Excel analysis
• Week 6: UK-domain project plus the personal site that ties it all together

This timeline assumes around 8-10 focused hours a week. It is the same pattern we see working for career changers - the same pattern described in our realistic Data Analyst timeline guide.

The Honest Summary

A great Data Analyst portfolio is not a wall of notebooks. It is four sharp projects, four clean READMEs, and one personal page that answers a recruiter's real question: can you do this job from day one?