Methodology

The Canadian Grocery Price Index (CGPI) is a weighted Jevons price index computed from a fixed basket of 50 essential grocery products. Prices are collected weekly from publicly available product listings on major Canadian retailer websites.

Unlike Statistics Canada's Consumer Price Index (CPI), which is published monthly with a 3-week lag, CGPI provides weekly price tracking with same-week publication.

Where:

• I_t — Index value at time t (base = 100)
• w_i — Normalized weight of product i in the basket
• P_it — Price of product i at time t (current period)
• P_i0 — Price of product i at the base period

The Jevons index uses a geometric mean (via log-transformation) rather than the arithmetic mean used in a Laspeyres index. This has two advantages:

1. Substitution bias correction — The geometric mean naturally accounts for consumers switching to cheaper alternatives when one product's price rises.
2. Symmetric treatment of price changes — A 50% price increase and a 50% decrease offset each other (they don't with an arithmetic mean).

This is the same approach recommended by the ILO/IMF Consumer Price Index Manual and used by Eurostat for elementary price aggregates.

Retailers

GroceryPulse tracks 23 banners across the three largest Canadian grocery parent companies plus two regional players.

Loblaw Companies

• Loblaws — Ontario full-service flagship
• No Frills — national discount banner
• Real Canadian Superstore — large-format (Western Canada and Ontario)
• Atlantic Superstore — large-format (Atlantic provinces)
• Dominion — Newfoundland full-service
• Maxi — Quebec discount banner
• Provigo — Quebec full-service
• Fortinos — Ontario full-service
• Wholesale Club — bulk / small-format
• Your Independent Grocer — smaller-market full-service

Metro Inc.

• Metro — Ontario / Quebec full-service
• Food Basics — Ontario discount banner
• Super C — Quebec discount banner

Empire Company (Sobeys)

• Sobeys — national full-service
• FreshCo — national discount banner
• IGA — Quebec full-service
• Safeway — Western Canada full-service
• Foodland — smaller-market full-service
• Thrifty Foods — BC full-service
• Farm Boy — Ontario specialty (via Voilà delivery)
• Voilà — Sobeys online grocery delivery

Independents / regional

• Save-On-Foods — Jim Pattison Group, Western Canada
• Giant Tiger — discount / general merchandise with grocery

Method

Prices are collected from publicly accessible retailer websites. Product listings are accessed via the same pages any consumer can view, and price data is extracted from the server-rendered HTML.

Product Matching

Each retailer's product listing is matched to our canonical basket using a Tier 2 specification matching system:

1. Type match — Product category must match (e.g., "milk", "butter", "chicken breast")
2. Size verification — Package size must be within ±10% of the target
3. Variant matching — Specific variants checked (e.g., "salted" butter, "whole wheat" bread)

The cheapest available match is used for the index, reflecting actual consumer behavior where shoppers tend to choose store brands over national brands.

Update Schedule

Data is collected every Thursday at 08:00 UTC (04:00 Eastern) via an automated pipeline. The index is recomputed after each collection run. A complete run covers all 13 cities and 23 banners; runtime is typically 10–30 minutes depending on retailer response times and rate-limit backoffs.

After every weekly scrape, an automated audit re-queries each retailer's live website for a stratified sample of stored prices and compares them to what we have on file. Audit verdicts are recorded in an internal scrape_audits table and form the basis of two quality controls:

1. Cent-level accuracy — fraction of audited prices that match the live retailer site to the cent (within 1¢ tolerance). Strict measure of read correctness.
2. Dollar-level accuracy — fraction of audited prices within $1.00 of the live retailer site. The gap vs cent-level is retailer-side price drift (sales rolling on or off between scrape and audit), not scraper error.
3. Auto-deactivation — any product mapping whose last two consecutive audits both fail (mismatch or not-found) is automatically retired from the index. The two-strikes threshold tolerates single-week sale rolloff while catching genuine defects within ~2 weekly cycles.
4. Mapping-defect bucket — audit verdicts with >30% delta between stored and live price are tagged as structural mapping defects (wrong product matched), not price mismatches, and feed the auto-deactivation pipeline.