https://github.com/obirikan/ad-performance-analysis
This project Compares Ad Effectiveness Using A/B Tests; analyzes ad performance using user interaction data, advertisement metadata, and device data. The goal is to evaluate click-through rates (CTR) across various ad versions, platforms, and devices.
https://github.com/obirikan/ad-performance-analysis
data-analysis pandas
Last synced: about 2 months ago
JSON representation
This project Compares Ad Effectiveness Using A/B Tests; analyzes ad performance using user interaction data, advertisement metadata, and device data. The goal is to evaluate click-through rates (CTR) across various ad versions, platforms, and devices.
- Host: GitHub
- URL: https://github.com/obirikan/ad-performance-analysis
- Owner: obirikan
- Created: 2025-06-03T16:30:48.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-06-09T19:33:45.000Z (about 1 year ago)
- Last Synced: 2025-06-17T03:09:09.379Z (about 1 year ago)
- Topics: data-analysis, pandas
- Language: Jupyter Notebook
- Homepage:
- Size: 249 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: readme.md
Awesome Lists containing this project
README
# Ad Performance Analysis
This project Compares Ad Effectiveness Using A/B Tests; analyzes ad performance using user interaction data, advertisement metadata, and device data. The goal is to evaluate click-through rates (CTR) across various ad versions, platforms, and devices.
## Dataset Overview
* `users.csv`: Contains user interaction data including user IDs, timestamps, device IDs, and whether an ad was clicked.
* `advertisements.csv`: Contains metadata about advertisements shown to users.
* `devices.csv`: Contains information about the device types used by users.
## Project Structure
The analysis is divided into five major task groups:
### Import, Inspect, and Merge
* Load CSV files using pandas.
* Inspect dataframes.
* Merge user and advertisement datasets on `user_id` and `timestamp`.
### Aggregations to Calculate Click-Through Rate
* Count ad views.
* Calculate unique users per ad version.
* Compute click-through rate (CTR) as a percentage.
### Comparing Ad Performances by Social Media Platform
* Analyze CTR across ad sources and versions using groupby and pivot tables.
### Comparing Ad Performances by Tech Device
* Merge in device data.
* Compare CTRs by device type and ad version.
### Weekday and Weekend Performance by Device Type
* Extract day of the week from timestamps.
* Compare performance metrics for weekdays vs. weekends.
## How to Run
1. Ensure the following dependencies are installed:
```bash
pip install pandas numpy
```
2. Place the `users.csv`, `advertisements.csv`, and `devices.csv` files in the working directory.
3. Run the Python script or load the notebook to perform the analysis.