Lever Factor brings causal attribution methods from academic research and big tech to ecommerce sellers who never had access to them.
Every ecommerce seller makes dozens of decisions every week. Change a price. Launch a campaign. Update a product photo. Reorder inventory. Try a new keyword.
And every week, revenue goes up or down. The question is: which decision caused which change?
The honest answer for most sellers is: they do not know. They have ad dashboards that show ad-attributed revenue. They have Amazon Brand Analytics that shows what happened. They have spreadsheets full of weekly performance data. But none of these tell them which of their actions actually moved the needle.
So they guess. They double down on what feels right. They keep spending on PPC because revenue is up, even when most of that revenue came from a price test they ran weeks earlier.
This costs money in two ways: spending on actions that do not work, and abandoning actions that actually do.
Lever Factor uses causal inference methods to build a counterfactual baseline for your revenue. The baseline answers a specific question: what would have happened if you had done nothing in this period?
Then for every action you logged, we measure how much of the actual versus baseline gap can be attributed to that specific action. The result is an LF Score from 0 to 10 with statistical confidence.
When the model cannot explain more than 10% of the change with your logged events, we call it a Hidden Lever. This tells you that something else happened that you forgot to log, or that the market moved in a way that affected your business.
We are a small team and we answer every message personally. Send us a note and we will get back to you within 24 hours.
Contact us