https://github.com/valenthr/ab_testing
Analysis of the AB Test Results for the Subscription Page Design
https://github.com/valenthr/ab_testing
Last synced: 10 months ago
JSON representation
Analysis of the AB Test Results for the Subscription Page Design
- Host: GitHub
- URL: https://github.com/valenthr/ab_testing
- Owner: valenthr
- Created: 2024-12-28T20:06:59.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-12-30T23:01:20.000Z (over 1 year ago)
- Last Synced: 2025-02-04T12:04:45.762Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 1.25 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# A/B testing
## Tools
Python libraries:
* Pandas;
* Scipy;
* Statsmodels;
* Matplotlib;
* Seaborn.
## Purpose
Make the right decision whether to apply a new design or not based on the results of an A/B test.
## Context of A/B test
In our mobile application, after onboarding, the user is offered to buy a weekly subscription for $4.99, which gives access to premium features.
### Groups
1. A - control group - old design with offer for the weekly subscription for $4.99.
2. B - test group - new design also with the same price, but we say it's actually a 50% off price.
## Data
All data is in table ['ab_test_data'](https://drive.google.com/file/d/11aIe3--lnKRPnerjnJEGloHSMwP3_rTz/view?usp=drive_link):
* user_id - unique id of each user;
* timestamp - time of displaying an offer of subscription;
* test_group - a or b;
* conversion - 0 if user rejected the offer and 1 if user buy the weekly subscription
## Analysis and final report
* Analysis with explanations - [there](https://drive.google.com/file/d/1mu6lFVQdXFCosVup2xnQWbOEEHVt6cTc/view?usp=sharing).
* Final report with conclusion - [there](https://docs.google.com/presentation/d/14YqG0EmGnjJcl9SFT2Cp-fn7fT1tkmSx/edit?usp=sharing&ouid=110123972342957476617&rtpof=true&sd=true).