In simple English
Online stores often test new product titles or descriptions to get more clicks. Guessing if a new title works better than the old one is hard without real tests. This tool splits your products into two groups so you can compare their performance fairly. It helps you see which changes bring more shoppers to your site. You need a bit of data knowledge and a Python environment to run it.
What it is
FeedX is a Python package and companion Google Colab notebooks for designing and analyzing A/B experiments on e-commerce product feeds. It uses statistical methods like crossover designs and CUPED adjustments to measure small performance changes accurately.
What it is good for
E-commerce merchants and digital marketers who want to test batch updates to product catalogs without guessing. It removes uncertainty when deciding whether new product titles or descriptions actually improve ad clicks.
- Implements advanced statistical methods like crossover design to improve test accuracy.
- Includes ready-to-use Google Colab notebooks for both experiment design and results analysis.
- Provides built-in simulation tools to estimate your minimum detectable effect before launch.
- Requires at least one thousand items in your feed for reliable test results.
- Needs historical Google Ads performance data to run the initial design simulations.
- Item-level randomization can inflate effect sizes due to product cannibalization.
Manual guesswork when updating product catalogs
Google Merchant Center
- jesse-ai/jesse8.7k stars
A Python framework for writing crypto trading strategies, testing them against past prices, and running them live.
- hummingbot/hummingbot20k stars
A free framework for writing and running automated trading programs on cryptocurrency exchanges.
- lballabio/QuantLib7.7k stars
The long-established free library for pricing and risk-managing financial instruments.
pip install feedx