Why restaurant owners discover cost drift too late

Owners discover drift too late because each signal lives in a different tool and comes out at a different pace. Prices are in invoices, waste in a logbook, hours in the schedule, sales in the POS. Nobody puts them side by side before the close, when it is already late to act.

The month that ends badly, with no one to blame

A fictional example. A group of three restaurants closes its month. The second site's margin is below expectations. The owner rereads everything: the chef changed supplier on two products, a price increase slipped through on an invoice, fish waste went up in the second week, and the manager added extra hours after an absence. Each of these causes was known to someone. None was known to everyone at once, or at the moment it began.

Why drift slips under the radar

• Signals are scattered. Prices, waste, hours, sales: four sources, four rhythms, four owners.
• Each cause is small. A rise on a few products, a slightly generous portion, one extra hour. Only their combination weighs.
• The close is the first moment when everything is put together. It is also the moment when nothing can be changed about the month.
• Decisions are not traced. An increase was accepted, but where is the reason written down? After three weeks, nobody remembers.
• The owner is not on site. In a multi-site group, they learn things from the result or from a phone call, never from the flow of the days.

From the close to the week: a method

1. Choose four facts per site, identical everywhere. For example: price gaps received on tracked products, waste with its cause, planned and actual hours, open incidents. The choice is made with the chefs and managers.
2. Set a weekly rhythm. One day, one time, one person in charge per site. Tool: a shared one-page template.
3. Write the reason next to each gap. Who, what, why, decided by whom. Without a reason, the gap is an anomaly; with one, it is a decision.
4. Ask three fixed questions every Monday. What moved? What has not been dealt with? What keeps coming back? The owner answers in ten minutes per site.
5. Decide on an action with a name and a date. Renegotiate, review a portion, train, reschedule.
6. Reconcile with the close. At the end of each month, compare the result with the gaps seen during the weeks. What was not seen is what you need to add to the tracking.

Where each signal lives, and when it comes out

Signal	Where it lives	When it comes out	What is missing
Price increase	Invoice, supplier price list	At the invoice or at the close	A comparison at receiving
Waste	Logbook or no record at all	At month-end stock	Cause logged on the spot
Extra hours	Schedule, payroll	At payroll	The reason for the gap
Sales by dish	POS	Immediate but not linked to costs	Link with recipes
Incidents	Messages, logbook	On request	History by site

Mistakes that delay the signal

• Waiting for a complete dashboard before starting to look.
• Multiplying indicators: beyond a few numbers, nobody reads.
• Writing down gaps without reasons.
• Correcting the person instead of looking at the process, the training or the schedule.
• Never going back over what the close had shown late.

In a multi-site group

With several sites, the delay multiplies: each one closes at its own pace and each manager keeps their own tracking habits. The owner then compares numbers built differently. Before comparing two sites, check that they count the same thing, the same way, on the same day. Otherwise the gap you see is that of the counting method, not of the operation.

What to measure

• The delay between the start of a drift and the day the owner sees it. To be reduced, measured in days.
• The share of gaps that have a written reason and a decision.
• The number of drifts discovered by the close rather than by weekly tracking. If it goes down, the tracking works.

Where Tsuno comes in

Tsuno keeps standards, observed facts and actions taken, dated and searchable, with their provenance. Instead of opening four tools, you ask: "What deserves my attention today?" or "How did the week go at site X?" The answer is short, based on the data, with the facts behind it. When data is missing, Tsuno says so. Your POS, purchasing and payroll tools stay in place, and their exports are imported with human validation. Tsuno prepares the action, you confirm. See the solution page.

Going further in this cluster

This topic breaks down into four readings: food cost that arrives late, the full cost of waste, the supplier price list that stays accurate and the arbitrary hours budget. Prime cost ties them together.

Key takeaways

Drift is not invisible, it is scattered. Choose four identical facts per site, read them every week with their reason, and the close stops being a surprise.