Asda Data Scraping Services

No member shelf price, so no effective-price problem

Tesco and Sainsbury's both run schemes that show a lower price to members on shelf. That creates the single largest analytical problem in UK grocery data: the displayed price is not what most baskets transact at, and an index built on it is wrong in a direction that flatters the retailer.

Asda's scheme works differently — value accrues to a wallet as cashback rather than reducing the price shown on the item.

Why that is useful rather than boring

• The shelf price is the price. No effective-price computation, no member-versus-standard split, no assumption about scheme membership.
• It is the clean baseline. Indexing Tesco or Sainsbury's against Asda isolates the loyalty effect, because one side of the comparison has none.
• Promotional depth is directly measurable. Depth against a previous shelf price means what it says.
• Cross-retailer indices are more defensible when at least one member of the set needs no assumptions.

We capture loyalty_mechanic as wallet_cashback on every record, so anyone joining this data to our Tesco or Sainsbury's feeds can see immediately which retailers need effective-price handling and which do not. That field exists precisely so the distinction cannot be lost in a join.

Rollback is a state with a duration

Asda's principal promotional mechanic is a sustained price reduction rather than a short multibuy. Treating each observation as a fresh price change misses what is actually happening.

• Duration matters more than depth. A rollback running eight weeks is a repositioning; one running one week is a promotion.
• Start and end dates make cadence measurable by category.
• Repeat rollbacks on one line indicate a price the retailer keeps testing at a lower level.
• The pre-rollback price is the reference, and it is observable because rollbacks have boundaries.

We capture is_rollback, rollback_since, rollback_days and rollbacks_last_12m. A line with six rollbacks in a year at similar depth is effectively repriced, and counting each as a promotion overstates promotional intensity — the same correction we apply to the Australian specials cycle.

Price matching claims need capture, not verification

UK grocers periodically make price comparison or matching claims against named competitors. Those claims are publicly displayed and are worth recording as data.

What we do not do is verify them. Verifying a price-match claim requires knowing the competitor's price at the moment the claim was made, on the exact matched product, under the matching scheme's own rules about pack size and promotional state. Those rules are not fully published.

• We capture the claim as displayed with a date, so a claim inventory exists.
• We capture our own observed prices at both retailers where we collect both.
• We do not compute a compliance verdict, because the scheme's matching rules are not public enough to apply.

You can compare our observed prices against the claim yourself and reach your own conclusion. That is a legitimate analysis. A vendor asserting a compliance rate is asserting a rules interpretation they cannot substantiate, which is the same reason we capture beauty claims without verifying them.