Woolworths Data Scraping Services

Mandated unit pricing lets you audit the arithmetic

Most retailers display unit prices inconsistently or not at all, which is why we compute our own on a stated basis. Australia is different: unit price display on grocery is a regulatory requirement, so a published figure exists for most lines.

Why we still compute our own

• Bases vary even where display is required — per 100g, per kg, per litre, per each — so cross-category comparison needs normalising regardless.
• Comparison catches errors. Where our computed figure and the displayed one disagree, that is worth flagging rather than silently preferring one.
• Some lines are exempt or display nothing, and those need computing.

We deliver unit_price_displayed, unit_price_computed and unit_price_agrees. That third field is genuinely useful: it lets you audit our arithmetic in this market and then trust the same method in markets where no published figure exists to check against.

We have not seen another vendor expose that comparison, and it is the cheapest possible credibility check on a dataset where unit price drives most conclusions.

The specials cycle has fixed boundaries

Australian grocery specials run on a weekly cycle with defined start and end days. That is a much more structured promotional rhythm than most markets, and it changes collection design.

• A capture is needed at each boundary, or you cannot tell whether a price was a special or a base price.
• Mid-cycle collection alone makes every price look like a special, since the majority of promoted lines are promoted for the full week.
• Depth is measurable cleanly because the base price is observable outside the cycle — a luxury most markets do not offer.
• Cycle-over-cycle repetition shows which lines are permanently promoted, which is a distinct commercial fact from a genuine promotion.

We guarantee a capture at each cycle boundary and record specials_cycle_id plus cycles_promoted_in_last_12. A line promoted in ten of twelve cycles is effectively at a lower base price, and treating each instance as a promotion overstates promotional intensity substantially.

A duopoly makes cross-retailer comparison the whole point

Australian grocery is dominated by two chains, which means almost every question a client has is comparative. That shapes what the dataset has to support.

• Product matching across the two needs to be reliable, since own-brand equivalents are not the same product but are the relevant comparison.
• Own-brand tiers must be classified per retailer, because tier positioning differs and a naive branded-versus-own comparison mixes tiers.
• Specials cycles may not align between retailers, so a same-day comparison can compare a promoted price to a base price.
• Postcode-level range differences mean a national comparison averages across different local ranges.

We deliver cross-retailer product matching with confidence, tier classification per retailer, and both_in_cycle so a comparison can be restricted to periods where both retailers' cycles align. Comparing a promoted price against a base price is the single most common error in Australian grocery analysis, and it is avoidable with one field.