Manage retail price and gross margin at the level of every SKU
One habitual markup costs money twice: where the price sits below the market you give away margin, and where it sits above you lose demand. This product measures both, line by line.
What is wrong today
- The markup is the same across a category, though price sensitivity differs line by line.
- Promotions run unmeasured: nobody knows what the discount delivered beyond baseline sales.
- Markdowns start late, once the goods have already lost value.
How it works
Each match to a market equivalent carries a confidence score. An unconfirmed match never feeds a recommendation.
A recommendation cannot fall below the threshold set for an item or a category — that is the network's rule, not the model's suggestion.
As the price moves, expected demand, gross profit and days of cover move with it on the same screen. The decision is made on all three at once.
What it looks like in use
One row is one item. The decision rests on three numbers at once: your margin, the market, and the risk of losing the shopper.
| Item | Price | Market | Margin | Suggested | Risk |
|---|---|---|---|---|---|
| Drinking water 0.5 l | 3.20 | 3.10 – 3.90 | 14% | 3.50 | low |
| Nourishing cream 75 ml | 34.90 | 29.00 – 38.00 | 41% | keep | none |
| AA batteries, 4 pcs | 18.00 | 15.40 – 17.20 | 26% | 16.90 | high |
| Shampoo 400 ml | 22.40 | 22.00 – 26.50 | 31% | 23.90 | low |
An anonymised example: locations are lettered and the dataset is shared across every screen on this site.
What it is built on
- Sales and cost per item are already read from the accounting system — actual margin is computed on them.
- Days of cover are computed today: without them, changing a price is blind — you can drive demand for goods you do not have.
- Approvals and the action log already work: the “proposed, approved by a person, recorded” mechanism is in place for orders.
Who stays in control
A price never reaches the till on its own. Every recommendation goes through approval, every change has an author, and write-back to the accounting system is enabled by a separate decision.
How it is measured
Through a control group: some items keep their price, and the effect is the difference in differences. Without a control group the product reports “not enough data”.
Show us one critical process. We will show how it runs here.
We look at your cycle: how an order is assembled today, who decides, where time leaks and what the system takes over.