Planning Produce for an Event: The Ordering Timeline

Say the week holds a 60-cover lunch on Tuesday and a 400-cover dinner on Saturday. Those are not the same service at a different size: they are separate orders, against different numbers, landing in different windows. That is the planning problem underneath a catering operation — volume spikes by the event, not by the night, so last week's order tells you little about this week.

Given a confirmed event, how do you get from covers to a produce order, and when does each decision lock. No published table of quantities will answer that; anyone handing you one is guessing at your portions and your prep. What exists is a method, and every number it needs is already in your own kitchen.

The timeline runs backwards from the event

Event ordering goes wrong when it is planned forwards: you start from today, work out what can still be done, and the order lands wherever the days run out. Planned backwards, the sequence is fixed, because each step gates the next. 5 gates, in this order: confirmed covers, then the menu, then the yield, then the order, then the delivery window.

You cannot set a quantity without a yield assumption, a yield without the cut and the treatment the menu calls for, or the menu without a cover count you trust. The last gate is the fixed point the rest is measured back from: each event takes a single contracted delivery window rather than a nightly run, so the window is the thing the whole sequence hangs off.

The event order is planned against the confirmed covers ahead of each event rather than left to the day it is needed — that is the difference between an event line and a standing one. The standing order cut-off is 23:00, with the order picked overnight; treat that as the door a late top-up still has to go through, not as the lead time an event runs on.

Write the gates down once, with your own number of days between each, and the ordering timeline stops being reinvented per event. Your own records set those numbers: line up the date the covers were confirmed against the date the order went in on the last 3 events, and the difference is the lead time the operation actually runs on.

Confirmed covers, not expected covers

Covers arrive in stages and only one stage is worth ordering against. An enquiry number is a sales figure and a provisional number is a hold; the number that belongs on a produce order is the guaranteed minimum in the contract, with a final number due by an agreed date — the way a function sheet is normally written.

So split the order rather than guess at the middle. The base goes in against the guaranteed minimum, because that is the number you are contractually going to cook; the balance goes in against the final confirmed number, on the lines where a late top-up is workable. Say an event carries a guaranteed minimum of 250 and a final number of 310: the base is built for 250 and the top-up for 60, rather than ordering 310 early and carrying the risk, or ordering 250 and finding the gap on the day.

The menu decides how forgiving the order is

The menu is a planning decision dressed as a creative one. Say the same 300 covers are written against 2 menus: the supply risk can differ completely, and the difference is visible long before anyone orders.

Ask 3 questions of every produce line as the menu is signed off. Is the line in season through the event week, or at the edge of its window where the grade moves. Is there a like-for-like the kitchen would accept if the market is short, and is that substitution agreed now rather than on the morning of the event. And does the line hold once delivered, or must it be dressed, cut or plated on the day.

No answer is wrong in itself, but a menu built on lines that hold buys a wider ordering window and room to absorb a late change.

Produce per cover: the calculation, not the table

The whole per-cover calculation is 4 steps:

1. Take the plated portion weight for the line, from your own plate-up.
2. Multiply it by the confirmed covers. That is the weight on the plate.
3. Divide by the yield fraction for that line. That is the weight to order.
4. Round up to the pack the line is sold in.

Worked through: say a dish plates 90 g of a leaf and the event is confirmed at 200 covers. That is 18 kg on the plate. If your own trimming records put the usable fraction of that leaf at 0.75, divide 18 by 0.75 and order 24 kg, then round up to the pack the line is sold in.

Neither input can be published by anyone but you. The portion is a decision your chefs make at the plate-up: weigh it once, on the actual plate, and it stops being an argument. The yield is a property of your prep and your spec: weigh the line as delivered, weigh the usable weight that leaves the bench, and the ratio between them is your yield for that line.

That is why a published table of grams per cover is worth less than an afternoon with a set of scales: a table cannot know whether your leaf arrives washed, or whether the portion is a garnish or the centre of the plate.

Yield is measured per line, never applied across the order

The most common error is one blanket allowance across the whole order. Trim loss is a property of the line, the grade and the season: a root that is peeled behaves nothing like a leaf that is picked over, and a line at the edge of its season does not trim like one at the peak. Run the test per line, record the result on your spec sheet, and re-check it when the season moves.

Any contingency on top of the calculation is a decision rather than a formula, and it should be visible. Put it on the sheet as its own named line instead of folding it into the portion weight — fold it in and the portion drifts upwards every time an event runs short, until nobody can say what the dish actually plates. Your own records size it better than a rule of thumb: compare what was ordered against what was used on the last 3 events.

Close the loop while the function sheet is still on the desk

The timeline only improves if the event is written up. Put 3 numbers beside each other for the lines that mattered: what was ordered, what was delivered, and what was used. That comparison is the only honest source for the next event's yield fraction and the next event's contingency, and it beats any published benchmark because it is measured on your prep, your portions and your suppliers.

Do it in the days after the event, not a month later. By then the delivery notes are filed, the covers number in memory has drifted towards whatever was actually cooked, and the event that would have calibrated the next one is gone.

Put the 5 gates on one sheet

Confirmed covers, menu, yield, order, delivery window — worked backwards from the event, with your own lead time between each. Every number the sequence needs is already in your operation: the portion is on the plate, the yield is on the scales, and the allowance is in the delivery notes and the waste log.

If your events calendar rebuilds its ordering timeline from scratch every time, book a call and we will walk through how the window, the list and the statement are set up for a catering operation.