If you’re learning data analysis, you’ve probably heard that “real-world practice matters more than theory.” And it’s true. Knowing SQL, Excel, or Python is great—but applying them to real data is where the Datasets real learning begins.
When I was first learning data analysis, it was hard to find interesting and useful datasets. So I compiled this list of free datasets that I used to hone my skills—and they can help you too.
Whether you are attempting to optimize SQL queries, create dashboards in Power BI or Excel, or simply play with data for the sake of fun, these datasets have got you covered from sales and marketing to HR and finance.
Let’s see 10 fantastic datasets you can download today and practice.
Table of Contents
Datasets for Data Analyst Practice
1. Marketing Campaign Dataset
This dataset is from a company that ran multiple marketing campaigns. It contains details about their customers, products, and campaign performance. If you’re interested in customer segmentation or marketing analytics, this is a great place to start.
Features :
- Includes customer demographics like age, income, and marital status
- Shows past campaign results (who responded, who didn’t)
- Useful for RFM analysis and customer segmentation
- Helps understand buying behavior and product preferences
- Good for visual storytelling using dashboards
- Great base for clustering and classification models
2. HR Analytics : Employee Attrition
Employee attrition is a major concern for HR departments, and this dataset gives you all the data needed to analyze it. You can explore why employees leave and what factors lead to longer retention.
Features :
- Employee details like department, education, and experience
- Attrition flag to identify who left and who stayed
- Perfect for correlation and regression analysis
- Helps you explore internal trends (e.g., promotions vs attrition)
- Great for HR dashboards and storytelling
- Ideal for building predictive models (classification tasks)
3. Online Retail Sales Data
This is a transactional dataset from an online store that sells gift items across Europe. It’s excellent for sales analysis, seasonality checks, and customer behavior trends.
Features :
- Contains invoice numbers, dates, product codes, and prices
- Tracks customer IDs and countries
- Great for analyzing repeat purchases and sales trends
- Useful for cohort analysis and retention metrics
- Ideal for Excel or SQL-based sales dashboards
- Can be used to build an RFM model or lifetime value calculator
4. Pizza Sales Dataset
This is a smaller and cleaner dataset, perfect for beginners to build confidence. It simulates a pizza shop’s daily sales and lets you create fun dashboards and visualizations.
Features :
- Includes order dates, pizza sizes, and toppings
- Lets you track quantity sold and revenue
- Easy to explore time-based trends (day of week, peak hours)
- Great for building Excel dashboards with pivot tables
- Fun way to learn data cleaning and summarizing
- Simple but rich enough to analyze business performance
5. Diabetes Health Indicators
If you’re interested in healthcare data, this dataset gives you access to real health indicators used to predict diabetes. It’s a good chance to work on binary classification models and health trend analysis.
Features :
- Contains variables like BMI, blood pressure, physical activity, and smoking
- Each row represents a patient with or without diabetes
- Great for logistic regression and model building
- Helps practice data cleaning (handling missing/null values)
- Ideal for creating informative health dashboards
- Useful for learning how to explain insights to non-technical stakeholders
6. NYC Taxi Trips
This public dataset includes millions of taxi ride records in New York City. It’s large, rich, and perfect if you want to try time series or geospatial analysis.
Features :
- Pickup and drop-off locations with timestamps
- Fare amounts, distance traveled, and payment types
- Can be used to find busiest times or most common routes
- Allows deep practice with datetime functions and aggregations
- Great for maps and visualizations using tools like Tableau or Python
- Good example of handling large datasets efficiently
7. Indian Government Open Data
India’s open data portal includes datasets from different ministries—education, agriculture, environment, transport, and more. This is perfect if you want to explore public sector data.
Features :
- Datasets available in CSV, JSON, and Excel formats
- Updated regularly and available in regional categories
- Can be used to analyze state-level trends or national programs
- Ideal for policy analysis or NGO-focused dashboard projects
- Good for working on data from emerging markets
- Helps practice cleaning semi-structured and government-formatted data
8. Stock Market Data – Yahoo Finance
Yahoo Finance offers free access to historical stock data. It’s a great option for learning time series analysis, trend tracking, and financial forecasting.
Features :
- Daily open, close, high, low, and volume data
- Data available for global companies, indexes, and ETFs
- Good for calculating moving averages, volatility, and returns
- Useful for portfolio tracking and investment dashboards
- Can practice API-based data pulls in Python
- Excellent for building time-based dashboards or reports
9. IMDb Movie Dataset
This dataset contains detailed info about movies and TV shows, including ratings, genres, and cast members. It’s great if you want to work with text data or build something creative.
Features :
- Movie titles, release years, genres, and user ratings
- Cast and crew details, box office data
- Great for sentiment analysis and trend discovery
- Helps practice string handling and filtering in SQL or Python
- Ideal for ranking movies or comparing across categories
- Fun project to display in a portfolio or blog
10. World Bank Open Data
World Bank’s open database contains global development data across countries. You can explore metrics like GDP, education, poverty, and health indicators over time.
Features :
- Country-wise data spanning decades
- Dozens of topics including finance, education, and energy
- Ideal for comparative country analysis
- Good for building Power BI or Tableau dashboards
- Helpful for academic research and presentations
- Great for time series and multi-variable analysis
Final Tips for Practicing Like a Data Analyst
Now that you have the datasets, here’s how you can make the most out of them:
- Pick a dataset you enjoy and start with a few simple questions.
- Define a goal : e.g., “Can I predict attrition?”, “What are the top sales months?”
- Use SQL, Excel, or Python to explore, clean, and summarize the data.
- Build dashboards or visual reports to explain your findings.
- Publish your work on GitHub, Notion, or your own blog to share with recruiters.
Working with actual real-world datasets is perhaps the best way to develop as a data analyst. Whether you’re a beginner or wanting to create a good portfolio, these 10 free datasets present just the right platform to practice using SQL, Excel, Python, Power BI, or Tableau. Ranging from sales and marketing information to health and finance records, every dataset presents its own challenges—and that’s what makes learning so fun and worthwhile.
Don’t simply download the data and read over it—play with it. Ask business questions, identify patterns, create visualizations, and attempt to solve actual problems. The more you play with the data, the more job-ready and confident you’ll be.
If you’re creating your data analyst portfolio, these datasets will assist you in building solid case studies that will impress recruiters. And if you’re getting ready for data analyst interviews, practicing with real data will hone your mind and problem-solving abilities.
Ready to take the next step? Pick any one of these datasets and start analyzing today. The journey from beginner to skilled data analyst starts with one dataset—and one good question.
