Automate restaurant supplier orders: complete 2026 guide

TL;DR. Automating supplier orders means calculating what you actually need from real sales and current stock — not from gut feel. Result: fewer stockouts, less overstock, and 2 hours less per week juggling lists. This guide covers the method, the tools, and the mistakes to avoid.

Context / Definition

Automating restaurant supplier orders means replacing the Monday-morning glance into the walk-in with a structured calculation: forecast sales × recipe card portion sizes − current stock = what you order. Nothing more. But that "nothing more" changes everything when you do it systematically, week after week.

Most operators still order on feel. They eyeball the walk-in, think back to last week, add a buffer "just in case." That's money thrown away twice: once in overstock that ends up in the bin, once in stockouts that cost you a missed dish and a disappointed guest.

How do you calculate supplier orders?

The answer fits in one formula. Forecast sales × recipe card portion sizes − current stock = quantity to order. Apply that to every line item, every week, and you stop ordering blind.

The formula in detail

Concrete example. You sell an average of 80 burgers a week. Your recipe card and portion sizes specify 180 g of ground beef per burger. You have 4 kg in stock.

80 × 0.180 kg = 14.4 kg to produce
14.4 − 4 = 10.4 kg to order

You order 11 kg (rounded up to supplier case size). Not 15 to be safe. Not 8 because "it'll be fine." 11 kg, calculated.

The 3 data points you need

1. Real sales — not a mental estimate. POS data from last week, or a 4-week rolling average.
2. Recipe card portion sizes — every ingredient, net weight after prep, for every dish. If your cards aren't current, the calculation is wrong from the start.
3. Current stock — weighed or counted, not guessed. A quick inventory before each order.

Why manual ordering costs you

Ordering on instinct means working with a permanent unknown. You never really know what you have in stock or what you'll sell next week. And you pay for that uncertainty twice.

On the overstock side

10% overstock on a £3,000-a-week order is £300 going in the bin or stuck in cash flow. Across 52 weeks, that's £15,600 of working capital tied up. It's not an outlier — it's the norm in any restaurant ordering without method.

Stock management for accurate ordering and order automation are two sides of the same coin: each feeds the other.

On the stockout side

A stockout is a menu item you have to "86" mid-service. It's the server apologising. It's the guest leaving with the impression of a place that's disorganised. And if it's your hero dish, it's a direct hit to average ticket.

The live supplier mercurial lets you see prices move in real time — and adjust order quantities if a SKU spikes that week. Without an up-to-date mercurial, you sometimes order at a higher price than you need to, without ever knowing.

Time wasted

Building an order by hand — calls to suppliers, scribbled lists, makeshift WhatsApp threads — averages 30 to 45 minutes per supplier per week. For a restaurant with 4 main suppliers, that's 2-3 hours a week burned on a task that can be automated.

Case study — Lunch Wagon, 2024

In 2023, when I took over Lunch Wagon in Albi out of administration, ordering was old-school. Look in the walk-in, estimate the week, ring the supplier. Result: weeks where we tossed ground beef because we'd over-ordered, others where we scrambled because we'd underestimated demand for an event.

In 2024 I switched to a system: orders calculated from the previous week's actual sales, cross-referenced with recipe cards and a fixed weekly inventory. The order generated automatically, the PO went straight to the supplier on WhatsApp.

Results over the next 6 months: -12% stockouts, -8% overstock, and 2 hours back every week on order prep. It's not magic — it's just method, applied.

What I wish I'd had at La Verrerie between 2015 and 2018 is exactly that: a system that calculates for me, sends the PO without me having to type it, and flags when a supplier price drifts. I'd have saved dozens of hours and probably avoided a few embarrassing stockouts mid-Saturday-night service.

The impact on food cost is direct and measurable: order tight, lose less, hold less, free up cash.

Walk-in glance, WhatsApp, or automated?

The "automated ordering" column assumes your recipe cards are current and you run weekly inventory. If those two conditions aren't met, automation just automates wrong data. A structured purchase order template is the first step before automating.

How to set up an automated ordering system

Here are the 4 steps in order. Don't skip — each one sets up the next.

Step 1 — Get your recipe cards current

Every menu item needs a recipe card with precise per-ingredient portion sizes. If a card is approximate, the calculated order is approximate. It's the foundation.

Step 2 — Build a clean mercurial

A mercurial is the list of every supplier SKU with its current unit price. Without a reliable mercurial, your software orders without knowing actual prices. The live supplier mercurial shows how to keep it current with low effort using invoice OCR.

Step 3 — Lock in a fixed weekly inventory

Same time, same day, same method. Every item weighed or counted. That's the "current stock" variable in the formula. Without it, you can't calculate what's missing.

Step 4 — Pick the right tool

A serious supplier order software 2026 needs to do at least four things: keep the mercurial current, cross-reference sales with recipe cards, generate the PO, and send it to the supplier. WhatsApp dispatch is now standard — 100% of Lunch Wagon orders went out via WhatsApp straight from the tool, no re-typing.

WhatsApp purchase orders: what to know

Sending POs over WhatsApp is now the most common practice in independent hospitality. Fast, traceable, accepted by virtually every supplier.

What many operators don't know: a WhatsApp or SMS message counts as commercial proof if it's preserved. In a delivery dispute — wrong quantity, price different from what was ordered — the screenshot stands up.

Best practice: always send your PO with the detail (items, quantities, units, expected price if known) and keep the conversation history. Some tools generate a PDF attached to the WhatsApp message — even better for traceability.

Common mistakes

• Keeping recipe cards rough. If the portion size on the card isn't what's used in the kitchen, the order calc is wrong from the start. A card refresh every 6 weeks is enough to stay reliable.

• Letting the mercurial drift. A supplier hikes prices in January and you keep ordering as if nothing changed. Result: food cost slides without you understanding why. Real-time price cascade is the only defence.

• Automating without validating. For the first two weeks of a new system, compare the generated order with what you'd have ordered manually. If the numbers diverge sharply, hunt the error in the cards or the inventory — not in the tool.

• Mixing units across suppliers. One supplier in 5 kg cases, another by piece, a third by the litre. If your tool can't handle purchase units that differ from production units, you'll order incoherent quantities.

• Ignoring order history. A 12-week history is what surfaces seasonality, events that drive demand, and suppliers who routinely under-deliver against the order.

Conclusion

Automating supplier orders means applying a simple formula — sales × portion sizes − stock — systematically, week after week. No magic. Method.

Three things to keep:

1. The formula only works if the data is clean. Recipe cards current, weekly physical inventory, real-time mercurial. Without those three pillars, you're automating mistakes.

2. WhatsApp purchase orders are valid and traceable. Keep your history, send full detail, and you've got real commercial protection.

3. Start small, validate, then scale. One supplier automated and audited for 3 weeks beats 4 suppliers half-automated and never checked.

Lunch Wagon got 2 hours back per week and cut stockouts by 12% in under 2 months after the switch. That's what method delivers when it's applied properly.

Last updated 2026. Written by Cyril Quesnel, founder of Onrush, chef and entrepreneur (La Verrerie 2015-2018, Lunch Wagon 2023-2026).