How we measure — and why we publish the bits that don’t flatter us.

Latest run

October 2026 — the same basket, four ways to buy it (three where ALDI doesn’t trade). Pick a region:

All regions at a glance — October 2026

The basket

Thirty-four items — the weekly mix of fresh, meat, dairy, pantry, frozen, cleaning and personal care. The same basket in every region. The composition is fixed: it only changes at an annual review, and every change is recorded in the log below.

Where the basket comes from: it’s built around the items WiseList users add to their lists most often — bananas, milk, chicken breast and their neighbours — not around items chosen to make any store look good.

The rules

Numbered so anyone — including the supermarkets — can cite exactly which rule they disagree with.

• Cheapest item meeting the spec, any brand.

  For every line we take the cheapest product in that store that satisfies the spec — own brand or name brand. That’s how price-conscious shoppers actually buy, and it’s the only fair way to compare ALDI’s private-label range against the majors.

• Sizes are normalised.

  Every line price is the price for the specified quantity: for items sold by weight, the unit price applied to the spec; for packaged items, the shelf price of the cheapest qualifying pack. Every line has a target pack size. We take the cheapest pack close to that size — inside the size window printed against each line in the basket table — at its shelf price. If a store only sells the item in a different size, we take its best-value pack between half and double the target (or, failing that, the best-value pack of any size) and convert the price to the target size using the unit price — and flag that line in the breakdown. Produce sold by the piece is priced by the piece; chicken is priced per kilo from the unit price; toothpaste, shampoo and toilet paper are priced per gram, per millilitre and per sheet, because pack sizes differ too much between stores to compare them any other way.

• Specials count.

  If an item is on special that week, the special price is used — that’s the price you’d pay. A member price (Woolworths’ “Was $X” tag with the free Everyday Rewards membership) counts the same way. Lines bought on special or at a member price are flagged in the breakdown.

• One store per supermarket per region, stated.

  Each region is priced at one store of each supermarket, in the metropolitan area named. Prices elsewhere in the state can differ, especially for produce.

  Region	Woolworths	Coles	ALDI
  Sydney NSW	postcode 2204 (Arncliffe)	postcode 2089 (Neutral Bay)	postcode 2204 (Arncliffe)
  Melbourne VIC	postcode 3147 (Ashburton)	postcode 3151 (Burwood East)	postcode 3147 (Ashburton)
  Brisbane QLD	postcode 4007 (Ascot)	postcode 4000 (Brisbane City)	postcode 4007 (Ascot)
  Adelaide SA	postcode 5061 (Unley)	postcode 5000 (Adelaide)	postcode 5061 (Unley)
  Perth WA	postcode 6008 (Subiaco)	postcode 6280 — Coles' WA-metro pricing zone	postcode 6008 (Subiaco)
  Hobart TAS	postcode 7005 (Sandy Bay)	postcode 7005 (Sandy Bay)	ALDI doesn’t trade here
  Darwin NT	postcode 0800 (Darwin City)	postcode 0830 (Palmerston)	ALDI doesn’t trade here

  ACT is not shown: in WiseList's weekly data the Coles and Woolworths ACT prices are copies of the NSW prices rather than separately collected, so a Canberra figure would be Sydney's number under another name.

• Prices come from the product, not a clipboard.

  Figures are drawn from WiseList’s own weekly price data — the same data the app shows shoppers that week, captured on the Wednesday the supermarkets’ specials change. Each published run is cross-checked against a logged Split by Store run in the app for Melbourne; the other regions are priced from the same data with the same rules. The test is the product doing its job in public.

• “Split by Store” means the cheapest two-store plan.

  The split total is what you’d pay buying each item at the cheaper of the two stores in the plan — the best of the possible pairs, the same way the app’s Split by Store works. Where ALDI doesn’t trade, the split is Coles against Woolworths. The three-store minimum is in the breakdown for completeness, but it isn’t the headline figure: three supermarkets in one week isn’t a practical shop. When the split isn’t meaningfully better than the cheapest single store, we print that.

• Gaps are filled by rule, never by hand.

  If nothing that meets an item’s specification is listed at a store in the test week, we use the price from the nearest week something was, and say so in the breakdown. Our weekly capture can also miss a single product at one store while the store still sells it. Where that product was in the store’s previous capture, most of the chain’s other capitals that listed it then still list it, at least two of those charged this store’s price last time, and the ones that did all charge one and the same price in the test week, we price the line from it at that price if that is cheaper than anything the store’s own capture holds, and say so in the breakdown. Where those conditions are not met, the line is priced from the cheapest product the capture does hold, with no mark. We never type a price in by hand.

• The basket is frozen.

  Item composition changes only at an annual review, with every amendment listed in the change log. We never adjust the basket to change a result.

What this test can’t tell you

• Your basket isn’t this basket — brand loyalties, dietary needs and household size change the result.
• Own-brand and name-brand products aren’t always equivalent in quality; Rule R1 compares prices, not taste tests.
• One store stands in for a whole city (Rule R4). Stock and produce prices vary between suburbs.
• Where an item wasn’t listed in the test week, the breakdown says which week its price came from, or that it came from the chain’s other capitals that week (Rule R7). A product our capture missed at one store, which that rule cannot confirm elsewhere, leaves the line on the next-cheapest product.
• The split assumes you’re willing to visit two supermarkets in a week. Petrol and time aren’t counted.
• A month is a snapshot. The archive below matters more than any single run.

Archive

One row per monthly run, for the region selected above. Full per-item breakdowns are downloadable so any figure can be checked. Runs before September 2026 were computed retrospectively from our archived weekly price data, under the same rules.

Change log

Every amendment to the basket or the rules, dated.

Corrections

On 7 October 2026 every month was re-priced after Rule R7 was extended (see the change log). 48 of the 8,398 store lines in the 13 months published to that date changed, each to a lower price. Every one is listed here with the price first published and the price now; the per-item CSVs carry the corrected lines.