Does Pantry read Coles, Woolworths and Aldi receipts?

Yes. Pantry reads a photographed receipt from Coles, Woolworths and Aldi, and from IGA and the corner grocer, through the same reader, because it reads the receipt itself rather than matching one supermarket’s template. The models behind that reader are trained on Australian receipt data in English, which is why Australia is where Pantry launches first. One photo of the docket, whichever chain printed it, and the whole shop becomes a dated list.

One reader, not three templates

The obvious way to build a receipt scanner is to teach it one supermarket at a time: where Coles puts the total, how Woolworths writes a quantity, what an Aldi line looks like. That approach breaks the day a chain changes its layout, and it never covers the fruit shop on the corner at all. Pantry is built the other way round. It reads the printed lines for what they are — a name, a quantity, a price — and works out the item from the shorthand, so a new layout is a new receipt to read, not a new template to write. Coles, Woolworths, IGA, Aldi and the corner grocer all go through the same way.

What the three dockets have in common

The differences between chains matter less than what every receipt shares. A supermarket receipt is a terrible document: names chopped to fit a narrow roll, quantities in more than one format, the occasional bare product code sitting where a tomato should be. Abbreviation differs between chains and sometimes between stores of the same chain. That is the problem Pantry is built around, and it is the same problem on all three dockets. The reader does not need to know which supermarket printed a line to work out that it is a two-litre milk.

What the receipt carries that a barcode cannot

A receipt is the only artefact from a shop that carries what you bought and when, for the whole basket at once. The loose carrots, the deli ham, the bakery loaf — none of them had a barcode to scan, and all of them are on the docket. The purchase date is printed at the top, and that is the day every shelf-life clock starts from: three days for the chicken, five to seven for the milk, per CSIRO. That is the whole reason Pantry starts from the receipt rather than the barcode.

The lines that come back wrong

Some will. Receipts are abbreviated, faded and inconsistent, and this site does not quote an accuracy number nobody can show the workings for. What Pantry promises is the recovery: you confirm the list after every scan, fixing a line takes two taps, and Pantry is built to remember the correction, so the shorthand that scanned strangely this week is recognised the next time it appears on a docket from that chain. Thirty seconds of review on shopping day is what makes every date on the list worth trusting for the rest of the week.

Photograph it on shopping day

Thermal print fades. The receipt from a warm car on Saturday is a different document by the following Friday, and the one at the bottom of a bag has been folded through most of its lines. The habit that makes the reader’s job easiest is the simplest one: photograph the docket when the shopping comes in, before it goes anywhere near a pocket. Non-food lines — dishwashing tablets, the bin liners — are filtered out on purpose. Pantry tracks grocery shelf life, not your whole shopping history.

The comparison worth running

A handful of other apps have started building toward receipt scanning, and how they describe their own is on this site. Pantry is on the waitlist, not the App Store, so nothing here is a head-to-head trial yet. When it ships, the test that matters is the one you run yourself: a docket from your own kitchen, from whichever of the three chains you shop at, photographed on your own phone. Join the waitlist to be one of the first to run it.