Churn Report Kit
Churn analysis with an actionable retention queue.
Validates customer data and produces risk scores and inspectable reports.
- For
- Customer success and marketing teams
- Engine
- Python · XGBoost
- Source code
- Open · MIT
- Buying
- Written, fixed-price pilot
What is Churn Report Kit?
Churn Report Kit is an open-source report pipeline that turns customer data into a churn analysis and a retention queue. It validates the data, scores risk and produces 21 files from one command, including predictions, calibration, a model card and a PDF report. It is not a live SaaS and claims no retention effect.
From one CSV to a ranked retention queue and 21 files.
- ✓ Pandera schema
- ✓ enough cases per class
- ✓ thresholds valid
- ! leakage warning: column created after churn
Visualisation. The file list is the README's; customers in the queue are representative and no model performance figure is claimed.
How it works
From the product's own documentation. Each number says where and on which version it was measured.
Data is validated
Pandera schema, class-count checks and leakage warnings for post-churn columns.
A model is trained
XGBoost by default, logistic regression as fallback. Probabilities are calibrated.
It is measured honestly
ROC and PR-AUC, top 10% and 20% capture, bootstrap confidence intervals.
A queue comes out
A ranked retention queue with risk, likely drivers and a recommended action.
- 21 files
- Predictions, calibration, model card, PDF report and manifest, from one command. Source
- Model card
- What the model knows and does not know, delivered in writing.
- Reproducible
- Config snapshot, output manifest and Dockerfile.
- Open source
- Runs end to end on the Kaggle Telco dataset in the public repo. Source
Stack
- Python
- XGBoost
- SHAP
- Pandera
- Pydantic
- Typst
- Docker
Try it
pip install -e ".[dev]"
churn-report data/telco_customers.csv --config examples/config.json --out outputs/telco_reportWhat it does not claim
It is a report pipeline, not a live SaaS, and it claims no proven retention outcome.
Current boundary
This is a report pipeline, not a live SaaS or proven retention outcome.
What a pilot covers
A sample starting scope. Final deliverables and acceptance criteria are fixed in the written proposal.
- Question
- Where is churn risk concentrated?
- Likely input
- Permitted customer history and a churn definition
- Deliverables
- Data validation summary
- Inspectable risk and segment report
- Draft action queue for follow up
Before you decide
Can I buy this product today?
A paid pilot and setup proposal can be prepared. Scope, data access, deliverables, timeline and a fixed price are agreed in writing after a discovery call.
Why is there no price?
Data volume, integration and success criteria differ from job to job. After the call you get a fixed-price written proposal; no paid work starts without your approval.
Do I need to upload live data here?
No. This site has no file upload or accounts. Data access, retention and permissions are set only in the pilot agreement.
What is the first step?
Choose Churn Report Kit in the form and describe the problem you need to solve. The data and pilot boundaries are then set together.
More in this discipline
- Vinctum AegisPython · XGBoost · SHAP
From churn risk to an economically grounded retention decision.
- LuceThemisRust · Python
Audits signal quality in quantitative research.
- Calybris CoreRust · Python
Decision infrastructure with replayable proof.