Driver-Based Revenue
Forecast revenue from the operational metrics that actually cause it — orders, customers, ad spend.
By default, revenue is projected from its own history — last month's total carried forward. That works for a business whose revenue is steady, and misleads every other kind, because a single total cannot tell you why it moved.
This extension computes revenue from the operational numbers that cause it instead.
The three common shapes
Why decomposing matters
A revenue total that falls 8% tells you nothing actionable. The same fall expressed as drivers does: orders flat and average order value down means a mix or discounting problem; orders down and average order value flat means a traffic or conversion problem. These have different fixes, and the total cannot distinguish them.
Each driver also projects on its own. Order volume and average order value rarely move together, so forecasting them separately and multiplying is more accurate than forecasting the product — and it stays accurate when one of them changes.
Reading it in the table
Driver-based lines expand in the table to show the drivers beneath them, and a formula button spells out exactly how the number was computed. If a forecast figure looks wrong, this is where you find out why rather than guessing.
Drivers are editable into the future, so they are also the levers the simulator moves.
The end-to-end version for shops is shop revenue forecasting.