Live analytics dashboard
Six headline KPIs plus interactive charts for department, city, tenure, age, work mode, overtime and satisfaction — all recalculated instantly as you filter.
Explainable people analytics
See who's at risk of leaving — before they do.
Workforce Attrition Intelligence turns raw HR data into live dashboards, an explainable risk score for every employee, and clear drivers behind each prediction — so your people team can act early, not react late.
6K+
Employee records modeled
44
Encoded model features
10
Interactive dashboard charts
5
Performance metrics tracked
Everything in one place
A complete attrition-intelligence toolkit
From board-level KPIs to a per-employee risk score, every view is interactive and connected to the same underlying data model.
Attrition risk scoring
Enter an employee profile and get an instant probability of attrition, banded into low, medium and high risk with a calibrated confidence gauge.
Explainable predictions
Every score comes with the factors driving it — standardized log-odds contributions show exactly what pushes risk up or down for that person.
Self-serve data explorer
Search, sort and paginate the full employee dataset. Bring your own CSV or Excel file and the whole platform re-computes around it — no code required.
Model performance built in
Accuracy, precision, recall, F1 and ROC-AUC, a confusion matrix and a ranked feature-importance chart keep the model transparent and auditable.
Runs anywhere, instantly
The analytics engine runs entirely client-side — no server round-trips, no external database. Deploy it once and it just works, even offline.
How it works
From spreadsheet to decision in four steps
No pipelines to configure and nothing to install — the whole flow happens in the browser.
1
Load your data
Start with the built-in HR dataset or drop in your own CSV / Excel export in the Data Explorer.
2
Explore the dashboard
Filter by department, city, gender and status to see attrition signals recomputed live.
3
Score an employee
Enter a profile and get an explainable probability of attrition with its top risk drivers.
4
Validate the model
Review accuracy, ROC-AUC, the confusion matrix and feature importance before you trust a score.
Transparent by design
An explainable model you can actually trust
Predictions come from a balanced logistic-regression classifier trained on encoded HR features. Because it's linear, every score decomposes into the individual factors behind it — no black box, no guesswork.
Inspect model performanceModel summary Trained & validated
AlgorithmLogistic Regression
Encoded features44
Train / test split80 / 20 stratified
Class weightingBalanced
ExplainabilityStandardized log-odds
Turn your HR data into early-warning signals today
Open the dashboard, filter your workforce, and score your first employee in under a minute. Everything runs right here in your browser.