A hiring manager reviews 200 applications for one data analyst role. Every resume says the same thing: SQL, Tableau, Python. No proof. No evidence. Just a list of tools that anyone could type.
Then one resume links to a GitHub portfolio with three real projects — messy datasets, documented data cleaning, and business recommendations that actually make sense. That candidate gets the interview. The other 199 don't.
A portfolio isn't a nice-to-have in 2026. It's the single most reliable way to prove you can do the job when you don't have the experience section to prove it. But most portfolio projects are terrible — and bad projects are worse than no projects at all.
How many portfolio projects does a data analyst need?
3–5 projects is the sweet spot. Fewer than 3 doesn't demonstrate range. More than 7 dilutes quality. The ideal portfolio has one beginner project (data exploration), two intermediate projects (multi-source analysis with dashboards), and one advanced project (end-to-end analysis with business recommendations). Quality matters far more than quantity.
What datasets should I use for data analyst portfolio projects?
Use real-world, messy datasets from sources like Census.gov, WHO, NYC Open Data, CMS (healthcare), or Kaggle competition datasets. Avoid tutorial staples like Titanic, Iris, and mtcars — hiring managers have seen them hundreds of times and they signal tutorial completion, not analytical thinking. The messier the dataset, the better — data cleaning is 60–70% of real analyst work.
Where should I host my data analyst portfolio?
GitHub for SQL scripts, Python notebooks, and project documentation. Tableau Public for interactive dashboards. A personal website or GitHub Pages for a portfolio landing page. LinkedIn for visibility. The minimum viable portfolio: a GitHub profile with 3–5 repositories and at least one Tableau Public dashboard. No personal website is needed to get hired, but it helps.
72% of hiring managers say a portfolio matters more than a certification. That number should change how you spend every hour of your preparation time.
Resumes describe skills. Portfolios prove them. For entry-level and career-change candidates, this distinction is career-defining.
A portfolio does three things a resume can't:
- Demonstrates process — not just "knows SQL" but how SQL is used to answer business questions
- Shows communication — a well-written README proves the ability to explain technical work to non-technical readers
- Proves initiative — building projects without being assigned them signals self-direction
Portfolios prove competence in a way that resumes and certifications cannot. A data analyst with 3 strong portfolio projects and no degree will outperform a candidate with a degree and no portfolio in most hiring processes.
Not all projects are created equal. Understanding what hiring managers evaluate separates impressive portfolios from forgettable ones.
Most portfolio projects fail for the same four reasons. Before building anything, understand what gets a project thrown in the "ignore" pile — because the mistakes are counterintuitive.
- Business question — A clear, specific question that a real company would ask
- Real dataset — Messy, multi-table, from a credible public source
- Data cleaning — Documented steps showing how the raw data was prepared
- Analysis — SQL queries, Python code, or both — with explanations
- Visualization — Charts or dashboards that tell the data story
- Findings & recommendations — What the data shows and what action it suggests
- Clean README — Professional documentation that a hiring manager can scan in 30 seconds
Every portfolio project needs a business question, a real dataset, documented data cleaning, analysis with visualizations, and written recommendations. The README is as important as the code — it's what hiring managers read first and often the only thing they read.
Here are 13 specific projects — organized by difficulty — that demonstrate the exact skills hiring managers evaluate.
Don't let "beginner" fool you — these projects, done well, have landed people their first $65K analyst roles. The bar isn't complexity. It's clarity, documentation, and a real business question answered with real data.
These projects demonstrate foundational skills: SQL querying, data exploration, basic visualization, and the ability to communicate findings clearly. Complete 2–3 of these before moving to intermediate projects.
E-Commerce Sales Analysis
- 10–15 SQL queries analyzing revenue by category, time period, and region
- 3–5 charts showing seasonal trends and category performance
- Written summary with 3 actionable recommendations
Public Health Dashboard
- Data cleaning notebook showing how CDC data was standardized
- Interactive Tableau dashboard published on Tableau Public
- Analysis comparing vaccination rates to hospitalization outcomes
HR Employee Attrition Analysis
- Exploratory analysis identifying top 5 factors correlated with attrition
- Comparison of attrition rates across departments, salary bands, and tenure
- Recommendations for HR retention strategies
City Bike-Share Usage Patterns
- SQL analysis of peak usage times, popular routes, and station utilization
- Geographic visualization showing trip density by station
- Recommendations for new station placement based on demand patterns
Personal Finance Spending Analysis
- Data cleaning script that categorizes raw transactions
- Monthly spending breakdown with trend analysis
- Dashboard showing spending patterns and savings opportunities
Beginner projects demonstrate SQL fundamentals, basic data cleaning, and the ability to answer a clear business question. Complete 2–3 beginner projects before moving to intermediate — they form the foundation of the portfolio.
Intermediate projects raise the bar: multi-source data, more complex analysis, and polished dashboards.
These are the projects that actually get people hired. Beginner projects show you can use the tools. Intermediate projects show you can think like an analyst.
These projects demonstrate deeper analytical thinking: combining multiple data sources, statistical analysis, and dashboard design that tells a complete story. They're the projects that actually get you hired.
Airbnb Market Analysis
- Multi-variable analysis of price drivers (location, property type, amenities, reviews)
- Neighborhood comparison dashboard on Tableau Public
- Statistical summary of which features most influence pricing
- Written recommendations for hypothetical new hosts
Customer Segmentation for E-Commerce
- RFM analysis (Recency, Frequency, Monetary) calculating customer value segments
- Customer segmentation with 4–6 distinct groups and behavioral profiles
- Dashboard showing segment characteristics and recommended marketing strategies
- Executive summary with 3 actionable recommendations per segment
Supply Chain Performance Dashboard
- Supplier performance scorecard (on-time delivery, defect rate, lead time)
- Geographic analysis of shipping delays by route and region
- Interactive dashboard tracking 8–10 supply chain KPIs
- Written analysis identifying top 3 bottlenecks with improvement recommendations
Social Media Engagement Analysis
- Engagement analysis by content type, posting time, and day of week
- Sentiment distribution analysis using basic text patterns
- Content strategy recommendations backed by engagement data
- Dashboard showing engagement trends and optimal posting windows
Hospital Readmission Risk Analysis
- Analysis of readmission rates by diagnosis, age group, and length of stay
- Correlation analysis identifying top risk factors for readmission
- Dashboard showing readmission patterns across hospital departments
- Recommendations for intervention programs targeting high-risk patients
Intermediate projects demonstrate the ability to work with complex, multi-source data and produce dashboard-level deliverables with business recommendations. These are the projects hiring managers weigh most heavily — they're closest to actual analyst work.
For candidates targeting senior-level or specialized roles, advanced projects demonstrate leadership-level analytical thinking.
One advanced project in a portfolio does more for credibility than five beginner projects combined. These are the projects that make hiring managers pause and think, "This person could start contributing on day one."
These projects signal senior-level thinking: end-to-end analytical rigor, sophisticated methodology, and the ability to drive business decisions. Include one advanced project to stand out in competitive applicant pools.
A/B Test Analysis Framework
- Statistical analysis: hypothesis test, p-value, confidence interval, effect size
- Sample size calculation and power analysis
- Visualization of conversion funnels for control and treatment groups
- Written executive summary explaining results in non-technical language
- Reusable A/B testing template in Python
End-to-End Business Intelligence Pipeline
- Python script that extracts, transforms, and loads data
- SQL views and aggregations for efficient dashboard querying
- Executive dashboard with 10+ KPIs, drill-down capability, and auto-refresh
- Technical documentation covering data flow, refresh schedule, and maintenance
Market Entry Analysis
- Multi-source data integration from 3+ public datasets
- Market scoring model with weighted criteria (population, income, competition, growth)
- Financial projection model in Excel showing revenue scenarios
- Executive presentation (5–7 slides) with market recommendation
- Complete analytical appendix with methodology and data sources
Advanced projects demonstrate what separates senior analysts from mid-level: the ability to define the question, integrate multiple data sources, apply rigorous methodology, and present findings at the executive level. One advanced project in a portfolio signals readiness for higher-level roles.
Projects are only valuable if hiring managers can find and evaluate them. Presentation matters.
A brilliant analysis buried in a poorly organized GitHub repo is invisible. Presentation isn't vanity — it's the difference between a hiring manager spending 30 seconds on your project versus 3 minutes. And 3 minutes is usually enough to earn an interview.
# [Project Title]
Business Question
[One sentence describing the problem this analysis solves]
Dataset
Tools Used
Key Findings
1. [Most important finding with specific number]
2. [Second finding]
3. [Third finding]
Recommendations
Files
Methodology
[2-3 paragraphs explaining approach, assumptions, and limitations]
- Every project has a clean README that a hiring manager can scan in 30 seconds
- SQL files include comments explaining business logic
- Jupyter notebooks have markdown cells explaining each analysis step
- At least one interactive dashboard is published on Tableau Public
- The GitHub profile README links to all projects with one-line descriptions
- 013–5 portfolio projects beat 10 certifications — portfolios prove competence, credentials prove course completion
- 02Use real-world, messy datasets from Census.gov, WHO, NYC Open Data, or Kaggle competitions — avoid tutorial staples like Titanic and Iris
- 03Every project needs: a business question, documented data cleaning, analysis, visualization, and written recommendations
- 04Beginner projects demonstrate SQL and data exploration; intermediate projects show multi-source analysis and dashboards; advanced projects signal strategic thinking
- 05Host on GitHub with clean READMEs and publish dashboards on Tableau Public — presentation quality matters as much as analytical quality
- 06The README is the most important file in each project — it's what hiring managers read first (and often the only thing they read)
Can I use Kaggle datasets for my portfolio?
Yes — but avoid the most common tutorial datasets (Titanic, Iris, mtcars, Boston Housing). Use Kaggle competition datasets or less common datasets from the Kaggle Datasets section. The key is choosing datasets that are messy enough to demonstrate real data cleaning skills and complex enough to support meaningful analysis.
Do I need a personal website for my portfolio?
No. A well-organized GitHub profile with clean READMEs and a Tableau Public profile with published dashboards is sufficient to get hired. A personal website adds polish but isn't required. If you build one, keep it simple — a landing page with project descriptions and links is enough.
How long should each portfolio project take?
Beginner projects: 1–2 weeks. Intermediate projects: 2–3 weeks. Advanced projects: 3–4 weeks. These timelines assume 10–15 hours per week of focused work. The most common mistake is spending too long on one project instead of building a portfolio with range.
Should I include group projects or only solo work?
Solo projects are stronger for portfolios because they clearly demonstrate individual capability. If you include a group project, clearly document your specific contribution — which analyses you ran, which dashboards you built, which sections of the report you wrote.
What if my portfolio projects use different tools than the job posting requires?
The analytical thinking transfers across tools. A strong Tableau project demonstrates dashboard design skills that apply to Power BI. A pandas analysis demonstrates data manipulation skills that apply to SQL. Focus on demonstrating analytical process and business thinking — tool-specific skills are the easiest part to learn on the job.
Prepared by Careery Team
Researching Job Market & Building AI Tools for careerists · since December 2020
- 01Storytelling with Data: A Data Visualization Guide for Business Professionals — Cole Nussbaumer Knaflic (2015)
- 02Occupational Outlook Handbook: Data Analysts and Scientists — Bureau of Labor Statistics (2025)
- 03Inside Airbnb: Adding Data to the Debate — Inside Airbnb (2025)