Stop guessing why the number moved
Decompose changes in any KPI into the underlying drivers, ranked by contribution — automatically.
faster than manual variance analysis at typical pilot customers
What's broken today
Variance reports tell you the number changed, not why.
Manual decomposition takes a week and an analyst.
Leadership wants the explanation in the meeting, not a follow-up email.
Built for the way operators actually work
Automatic variance decomposition
Pick a KPI and a comparison period — we'll rank what drove the change.
Driver importance with statistical rigor
Shapley-style attribution, not just correlation. We show contribution and confidence.
Plain-English explanations
Every driver comes with a one-sentence explanation a non-analyst can use in the meeting.
Ask Predict Labs anything
Three steps to your first prediction
Pick the metric you care about
Revenue, churn, margin, conversion — anything in your warehouse.
Choose what to compare
Quarter-over-quarter, vs. plan, vs. forecast, vs. last year.
Read the answer
Ranked drivers, contribution percentages, and a short narrative summary.
What you'll need
These are the typical data sources for key drivers analysis. You don't need every source on day one — start with what you have.
- The metrics already in your warehouse
- Dimension tables (region, product, segment)
- No new data required
Try Key Drivers Analysis on your own data
Free for 14 days. No credit card. No data scientist required.