Hi everyone,

I recently worked on an A/B Testing project where I analyzed user behavior to determine whether a new version of a product performs better than the existing one.

What the project does

• Performs A/B testing on user data
• Compares control vs treatment groups
• Calculates statistical significance
• Helps decide whether to adopt a new feature

Why I built it

I wanted to strengthen my understanding of hypothesis testing and real-world data analysis, especially how companies use experiments to make product decisions.

Tech stack

• Python
• Pandas
• NumPy
• Matplotlib / Seaborn
• Statistical testing (hypothesis testing, p-values)

Key features

• Data cleaning and preprocessing
• Conversion rate comparison
• Statistical test implementation
• Clear visualizations for insights

GitHub Repo

https://github.com/gagan-gag/A-B-Testing

https://preview.redd.it/4mr0xtqpf4yg1.png?width=1366&format=png&auto=webp&s=e30958ddfcb48c3b2023bd5f272f4f6ee8890b5f

https://preview.redd.it/i4072uqpf4yg1.png?width=1366&format=png&auto=webp&s=c83226d351975bde07a780e5440f6c881d1eee52

https://preview.redd.it/82nr8uqpf4yg1.png?width=1366&format=png&auto=webp&s=0e337cbe7b9e368516e943c45f589d2ba8625d49

https://preview.redd.it/jamn3vqpf4yg1.png?width=1366&format=png&auto=webp&s=ee2149f6eea01c8ab641ff0d0aac446039c2f40b

https://preview.redd.it/c9ir0wqpf4yg1.png?width=1366&format=png&auto=webp&s=5a269a0b22f45df4dd32b7227284db81ff3f6c06

https://preview.redd.it/qih26wqpf4yg1.png?width=1366&format=png&auto=webp&s=d92721e5df7715dd6dc15cdc0f6524e09404d2fa

https://preview.redd.it/h72bzvqpf4yg1.png?width=1366&format=png&auto=webp&s=797f5e3b24d57c72243cd770dd004a2ac47af885

https://preview.redd.it/x9vaiwqpf4yg1.png?width=1366&format=png&auto=webp&s=7b225f45d4482c3360ec5369bc783d620a675e91

https://preview.redd.it/2uhcvwqpf4yg1.png?width=1366&format=png&auto=webp&s=c1e106aaec0e7e72e9e0ca1b5b132d434fe7d8c3

I’d really appreciate feedback on:

• Code structure
• Statistical approach
• Improvements or optimizations

Thanks!