Data Analyst Portfolio Tips for Complete Beginners in 2026

Career Guide · Data Analytics

Data Analyst Portfolio Tips for Complete Beginners in 2026

SECTION 01The 4 project types

A strong data analyst portfolio shows breadth without showing redundancy. You don't need 20 projects — you need these four types, done well.

SECTION 02How to structure your GitHub

Your GitHub is your portfolio. It should look organized, not scattered. Here's a simple structure that works:

your-github/
├── README.md  ← your portfolio homepage
├── project-01-sales-analysis/
│   ├── README.md
│   ├── data/
│   ├── notebook.ipynb
│   └── output/charts/
├── project-02-hr-dashboard/
│   ├── README.md
│   ├── dashboard.pbix
│   └── screenshots/
├── project-03-sql-queries/
│   ├── README.md
│   ├── queries.sql
│   └── results/
└── project-04-end-to-end/
    ├── README.md
    ├── notebooks/
    ├── app/
    └── screenshots/

// Naming conventions that work
// Good:
// - "sales-analysis-retail-2026"
// - "hr-attrition-dashboard"
// - "ecommerce-cohort-analysis"

// Bad:
// - "project1"
// - "test"
// - "data-analysis" (too generic)
// - "netflix-dataset" (done by 10k people)

// Prefix with type when relevant:
// - analysis-XXX
// - dashboard-XXX
// - sql-XXX
// - e2e-XXX

// Each repo should be:
// - Public
// - With a real README
// - With a clear purpose in one sentence

SECTION 03How to write a README that works

The README is the first thing a recruiter reads. Most beginner READMEs are just a title and a photo. Yours shouldn't be. Use this template:

# Project Title

## The question
What business question does this project answer?

## Data
Source, size, time period, notes on cleaning.

## Approach
Steps you took — tools, methods, decisions.

## Key insights
3–5 bullet points with the actual findings. Use numbers.

## Charts
Screenshots of 2–3 key charts.

## What I'd do next
Honest reflection — what would you improve?

## How to run
Brief instructions to reproduce the analysis.

# Retail Sales Decline — Root Cause Analysis

## The question
Why did Q3 revenue drop 12% vs. Q2 despite
higher web traffic?

## Data
- Source: Kaggle "Retail Transactions 2025"
- 480k rows, 12 months
- Cleaned 8 issues (duplicates, date formats, outliers)

## Approach
1. Explored trends across regions and categories
2. Segmented users by acquisition channel
3. Compared conversion by product category
4. Cross-checked with support ticket data

## Key insights
- Conversion dropped 18% on mobile (not desktop)
- Mobile checkout failed 6% of the time
- Payment success dropped from 94% to 88%
- Affected mostly new users, not repeat buyers

## Charts
- fig1-conversion-by-device.png
- fig2-payment-success-trend.png
- fig3-new-vs-repeat-users.png

## What I'd do next
- Pull in gateway logs to pinpoint payment failures
- Segment by OS version

## How to run
`pip install -r requirements.txt && jupyter notebook`

SECTION 04The 90-day plan

Here is a realistic 90-day plan for a complete beginner. Adjust to your pace, but keep the sequence.

SECTION 05Common mistakes

Avoid these — they're what kill otherwise good portfolios:

• Using only Titanic/Netflix datasets — everyone does these. Pick something with a business angle.
• No README — a repo without a README is invisible to recruiters.
• Hiding insights in the notebook — put your findings on the README, not on page 34 of a notebook.
• No charts — a data project without visuals is a red flag.
• Repos with old/stale code — if you're applying today, your recent work should be your best work.
• No live dashboard link — link to a public dashboard if possible. It's 10x more impressive than a screenshot.

SECTION 06Interview Q&A — portfolio tips

SECTION 07Test yourself — portfolio tips quiz

SECTION 08Frequently asked questions

SECTION 09Related reads