What are the three figures?
Cash sitting still is the value of stock that is not moving at its current rate of sale, computed per line from units on hand, cost and days of cover. Cash leaked in the last thirty days is money that left the business without a corresponding gain: realised discount given to customers who were buying anyway, the profit lost to returns, and sales lost in busy hours the store could not serve.
Cash at risk if nothing changes is the forward view: lines that will cross into dead stock within their window, seasonal stock that will miss its season at the current pace, and margin that will leak if a discount code keeps running. All three are dollar figures computed by the analytics engine, not estimates written by an agent.
Why does the strip sit above the KPI tiles?
Because revenue is the number a store admin already shows, and it is not the number an owner needs first. A month of strong revenue can hide a quarter of stock going dead and a code leaking margin every day. The strip is designed so the first thing on the numbers page is the money that is actually at stake, before the comparison deltas and the charts.
It also sets the order of the page. The KPI tiles, the net-sales bridge and the what-changed-and-why decomposition below all explain the strip's figures in more detail, so reading the Dashboard top to bottom moves from what matters to why it moved.
How each figure is computed
Cash sitting still: for each variant, units on hand multiplied by cost, included when days of cover exceed the threshold the owner set for the category. Cash leaked: realised discount on orders in the window, minus the discount that is judged to have driven incremental purchase from the per-code before-versus-during verdict; plus profit lost to returns; plus lost sales computed from footfall conversion in busy hours where people counting is connected.
Cash at risk: lines projected, at the current rate of sale, to reach the dead-stock window or to miss a season end date, valued at cost. All three are set-based queries over the engine's own database, unit-tested, and identical on every page that cites them. Where an input is missing, the strip says so rather than showing a smaller number.
Does every figure drill to the lines behind it?
Clicking cash sitting still opens the inventory analyst's stuck list, ranked by money, with the send-to-manager workflow already attached. Cash leaked opens the discount monitor and the returns analytics, each ranked the same way. Cash at risk opens Foresight's stockout and dead-stock predictions.
The drill-down is what makes the strip a decision tool rather than a headline. An owner who sees a large figure for cash sitting still can reach the top ten lines behind it in one click, propose a transfer or a markdown for each, and send the plan to the merchandiser through Review & Send.
What the strip says when the data is partial
In the first weeks, cost coverage may be partial, so cash sitting still describes only the lines with a cost on record and says so. Where people counting is not connected, lost sales in busy hours are not reported and the leaked figure carries the note. The strip never fills a gap with an estimate.
This honesty is deliberate: an owner who reads a strip figure as complete when it is not would make a wrong decision with confidence. As costs are loaded and integrations are added, the coverage rises and the strip tells the owner it has.
Which roles work from the strip on ftes.ai?
The Cash Officer FTE owns cash sitting still and cash at risk in stock; it is judged on cash released from inventory. The Pricing Officer FTE owns the discount part of cash leaked, and the Returns Officer FTE owns the returns part. The Store Officer FTE owns lost sales in busy hours where footfall is connected.
Each role reads the strip's figure for its area, proposes the actions behind it, and hands them to the owner through Approvals. The strip is the shared scoreboard; the roles are the people accountable for moving it.