Upload your sales history and get a straight answer for every product — how much to order, what is about to run out, and where your money is stuck. No spreadsheets to build, no analyst to hire, no software to install.
Four questions every business with stock has to answer, and usually answers by instinct.
Products heading for zero, ranked by what the stockout would cost you — before a customer walks to a competitor.
Stock that isn't moving, with a figure attached. Most businesses are surprised by how much is sitting there.
A quantity for every product — with the method used and how far to trust it stated plainly, not hidden.
Who actually delivers on time. Unreliable suppliers quietly force you to hold extra stock to cover them.
Export from your till, accounting package or a spreadsheet. The app works out which columns are which.
Before showing a single number, it tells you whether your data can actually support a forecast — and what it can't.
A short, prioritised list of what to do, with the money involved beside each one.
This is not a claim about our policies — it is how the software is built. Your file is read and analysed entirely inside your own browser. There is no server to send it to.
No account. No upload. No cookies. No tracking. You can disconnect from the internet after the page loads and everything still works.
A realistic two-year sample: 15 products with genuine seasonality, a few irregular sellers, some returns and duplicate rows, and stock levels that are deliberately wrong in places — so you can see what the app catches.
Download all four files, or just the first — the rest are optional.
Stock levels · Costs · Purchase orders
Or click Try it with demo data inside the app — no download needed.
| You need | Example |
|---|---|
| A date | 2024-03-15 |
| A product name | Cotton T-Shirt White |
| A quantity sold | 47 |
That's the minimum. Add stock levels, costs or purchase orders and more of the app unlocks. Full guide →
We're adding features based on what real businesses ask for. Drop us a line and we'll keep you posted — no marketing lists, no automated mail.
This opens your own email app — nothing is submitted from this page. Skip and open the app →
Implements the no-code AI inventory method of Jauhar et al. (2024, International Journal of Production Research), extended with Croston's method and the Syntetos–Boylan Approximation so products that sell irregularly are handled properly rather than ignored.
Unusually, the app measures itself against a naive benchmark on your own data and tells you when the forecasting is not adding value. Published studies compare AI against other AI; they rarely ask whether it beats simply using each product's average.
Read the method →