https://github.com/jroakes/npath
Exploring path sequences in GA4 BigQuery data
https://github.com/jroakes/npath
analytics bigquery pathfinding-algorithm
Last synced: 2 months ago
JSON representation
Exploring path sequences in GA4 BigQuery data
- Host: GitHub
- URL: https://github.com/jroakes/npath
- Owner: jroakes
- Created: 2023-10-24T11:23:25.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2023-11-27T14:46:46.000Z (over 1 year ago)
- Last Synced: 2025-04-09T15:10:46.523Z (2 months ago)
- Topics: analytics, bigquery, pathfinding-algorithm
- Language: Python
- Homepage: https://locomotive.agency
- Size: 182 KB
- Stars: 3
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: Readme.md
Awesome Lists containing this project
README
# NPath
[](https://colab.research.google.com/drive/1BKrdrLrWdxUZFPnSxUJWZW4wfoulavUx?usp=sharing)
## Description
Exploring path sequences in GA4 BigQuery data
## Setup
1. Create a new Google Cloud project
2. Enable the BigQuery API
3. Ensure that GA4 data is being sent to BigQuery
4. Get the dataset ID and table ID for the GA4 data
5. Create a service account with BigQuery read access. [Here](https://docs.aws.amazon.com/dms/latest/sbs/bigquery-redshift-migration-step-1.html) is a good guide.
6. Download the service account key as a JSON file
7. Create a new file in the root directory called `service_account.json` and paste the contents of the JSON file into it
8. Run `pip install -r requirements.txt` to install the required Python packages
9. Get your API key from OpenAI if you want to run analyze_clusters.
10. Open `demo.ipynb` in Jupyter Notebook and run the cells
## Components
* `plot_important_features_prefixspan`: Plots the important conversion path sequences of the PrefixSpan model.
* `convertor_review`: Sequence Patterns of Similarity and Anomalies in Non-Convertors that are clustered with Convertors
* `analyze_divergence`: Scores the similarity of non-convertor to convertor sequences.
* `analyze_clusters`: Clusters users based on their navigational paths and labels clusters.
## To Do
- [ ] Add more documentation
- [ ] Update sequence importance for sequences that pass through certain pages.
- [ ] Add more sequence mining algorithms
- [x] Add attribution models
- [x] Remove sequences after conversion
- [x] Add sequence divergence
- [ ] Analysis by section
- [ ] Analysis through product/service page
- [ ] Analysis through blog
- [ ] Analysis through pricing page
- [ ] Analysis by source/medium
- [ ] Analysis to score pages based on their importance (presence in conversion and closeness to conversion)