Looking in the fridge and hoping
On Wednesday afternoon you stand in the walk-in with a notepad. How much chicken for the weekend? Last weekend you ran out of breast by nine on Saturday. The weekend before, you binned a tray of marinated thigh on Monday. The wholesaler's rep needs the order by four. You write down roughly what you ordered last time, add a bit because the weather looks good, and hope.
The problem is not that you are bad at it. You are probably quite good at it. But the information you would need, how many of each dish sold on each channel on each day, is locked in three tablets and a till, and nobody has time to add it up.
Why ordering stays a gut call
Your sales data exists, but it is split. The till has counter and phone sales. Each delivery app has its own order history in its own portal. The website has its own. To know how many chicken dishes sold last Saturday you would need to export from four places and add them up by dish, and dish names are not even spelt the same on every app.
Even with the dish counts, you need to know what goes into each dish: grams of chicken in a main, in a wrap, in the mixed grill, in the family box. That is in the chef's head, not written down. So the link from sales to stock is never made, and the order is guesswork informed by experience.
The cost of guessing
| Result of guessing | What it looks like |
|---|---|
| Running out | Dishes switched off on a busy night, lost orders |
| Over-ordering | Waste binned early in the week |
| Emergency trips | Someone sent to the cash and carry mid-service |
| Price surprises | Ordering at whatever price the rep quotes that week |
| Owner dependence | Only one person can place the order |
Waste and run-outs both hit margin, and in a takeaway the margin on a dish is not large. The weeks you get it right are luck as much as judgement.
From dish sales to a suggested order
- We combine dish sales from your till, each delivery app and your website, matching the same dish across channels even where names differ, using the master menu if you have one.
- With the chef, we write down the main ingredients for each dish in simple quantities. It does not need to be a full recipe, just the items that drive the order: meat, fish, rice, bread, oil, packaging.
- Sales for the coming days are estimated from the same days in recent weeks, adjusted for anything you flag, such as a bank holiday or a big match.
- Current stock is entered from a quick count on a phone, only for the key lines, so the tool knows what is already on the shelf.
- The tool produces a suggested order per supplier, grouped the way each supplier sells, with the reasoning shown next to each line.
- You adjust and approve it, and it can be sent by email or placed in the supplier's portal where they have one.
The suggestion is only as good as the recipe quantities and the stock count, so we keep both short and practical. Your judgement stays in charge of the final order.
Wednesday afternoon, done differently
You open the suggested order on your phone. It says the weekend usually needs this much chicken breast based on the last four weekends, less what you counted this morning, plus a note that last Saturday you switched breast off at nine. You add a little for the forecast and send it. On Monday the waste bin has less in it.
- Orders based on what really sold on every channel
- Fewer run-outs on busy nights
- Less waste binned at the start of the week
- Someone other than the owner can place a sensible order
Does your stock ordering work like this?
- You order by looking in the fridge and remembering last week.
- You have run out of a key ingredient on a busy night recently.
- Good food gets binned because too much was ordered.
- Sales from the apps are not included in any stock calculation.
- Only one person knows how to place the weekly order.