In an independent kitchen, deliveries are checked against a paper docket by whoever is nearest the back door during service. Nobody transcribes those dockets. So a short delivery goes unnoticed, a supplier price rise goes unchallenged, and the only signal that anything is wrong is a gross-profit number that drifted and nobody can explain.
Full restaurant inventory platforms solve this and charge per location for recipe costing, supplier catalogues and ordering. That is worth it at scale. For one site with a tight margin, the realistic choice is between a clipboard and a spreadsheet.
Log the delivery at the door
This is the highest-value scan in the whole operation, because it is the only moment you have leverage. Once the driver has gone, a missing case is your word against the docket.
- Scan each case or product code as it comes off the van.
- Note the quantity against what the docket claims, and raise it there and then.
- Record who received it — accountability changes behaviour more than any report.
- Photograph damaged goods before they go in the walk-in.
- Log the delivery even when it is correct. Only a complete record proves a pattern.
Back doors and walk-in fridges are signal dead zones almost without exception. Anything that submits over the network at the moment of scanning will fail there. Test in the actual doorway, not in the office.
The free route
A Google Form linked to a Sheet, opened on a phone at the back door, costs nothing. For a site taking three deliveries a week it may be entirely sufficient, and we would rather say that than pretend otherwise.
- Free, no cap, nothing to buy.
- Fails in the dead zone by the back door, which is where you need it.
- Slow if a delivery has thirty lines and each needs a form submission.
- No accountability unless whoever is receiving remembers to enter their name.
Where a scanner link helps
We build one, so read this as interested and specific. Two things matter in a kitchen. First, scans are saved to the device before anything is sent, so the back door and the walk-in stop being a problem. Second, there is nothing to install and no account for staff — which matters in an industry with high turnover, where a new kitchen porter needs to be useful on their first shift, not enrolled in software.
Pricing shape: QR to Sheets is free for 300 scans, one link and one device, then $7 per registered device per month with unlimited scans. One phone by the back door is one device. Nobody pays per member of staff, which is the right shape when the rota changes weekly.
| Capability | Google Forms / Apps Script (DIY) | QR to Sheets |
|---|---|---|
| Cost | Free (Google Form) | Free to 300 scans, then $7 per device per month |
| Works at the back door | No — signal dead zone | Yes, saved on device |
| New staff onboarding | Explain the form | Open the link |
| Per-staff cost | None | None — priced per device |
| Who received it | Only if typed | Recorded per device |
| Recipe costing and ordering | No | No — use an inventory platform |
| Where data lives | Your Google Sheet | Your Google Sheet |
Weekly counts and the variance that matters
Deliveries tell you what came in. Counts tell you what is still there. The gap between them, set against what you sold, is the only honest measure of waste, over-portioning and loss.
- Count the same shelves on the same day each week. Consistency beats completeness.
- Count high-value items weekly and everything else monthly.
- Investigate variance on spirits, protein and anything easily pocketed first.
- Expect variance. A number that is exactly zero every week means somebody is copying last week's figures.
What this does not do
- No recipe costing or theoretical yield.
- No supplier catalogue, ordering or invoice matching.
- No allergen or HACCP documentation — those have their own legal requirements.
- No integration with your POS, so the sales side of the comparison is manual.
If you need those, buy a hospitality inventory platform. This gets you a trustworthy record of what arrived and what is on the shelf, which is the input every one of those systems depends on anyway.
Step by step
- 1
Log deliveries as they land, not later
Scan at the back door while the driver is still there. A discrepancy found at the point of delivery is a credit note; the same discrepancy found at close is an argument you will lose.
- 2
Scan what the supplier already labels
Case barcodes and supplier product codes are already on the box. Use those before printing anything of your own.
- 3
Record the count, not just the arrival
A scan says the case arrived. A scan plus a quantity says whether the order was complete. The second one is what saves money.
- 4
Do a weekly count on the same day
Consistency matters more than frequency. The same shelf, the same day, the same person if possible — variance between counters looks exactly like theft in the data.
- 5
Compare purchases against sales monthly
Delivery logs on their own are admin. Set against sales, they show yield problems, over-portioning and waste, which is where the margin actually is.
Frequently asked questions
Why log deliveries if the supplier provides a docket?+
Because nobody transcribes dockets, so patterns stay invisible. A logged delivery lets you show a supplier that three of the last eight orders were short, which is a conversation. A drawer of paper dockets is not.
Will scanning work by the back door with no signal?+
Only with a tool that saves each scan on the device and syncs later. A Google Form submission will simply fail there. Back doors and walk-in fridges are reliably dead zones, so test in the actual doorway before committing.
Do kitchen staff need to install anything or have accounts?+
With a scanner link, no — they open a URL in the phone browser. That matters in hospitality, where turnover is high and a new starter needs to be useful immediately rather than enrolled in software.
Does this replace a restaurant inventory system?+
No. There is no recipe costing, supplier ordering, invoice matching or POS integration. It gives you an accurate intake and count record — which is the data those platforms need as input, and the part most independents are missing.
How often should we count stock?+
Weekly for high-value items — spirits, protein, anything easily pocketed — and monthly for the rest. Consistency matters more than frequency: the same shelves, the same day, ideally the same person, because variance between counters is indistinguishable from loss.
What is the single most useful column to add?+
Who received the delivery. Accountability changes behaviour faster than any report, and it takes one extra scan or one separate link per receiving station to capture.