How to Scrape Save-On-Foods, PriceSmart and Urban Fare Prices

Start here

Paste this into the Input tab of the Save-On-Foods Scraper API, switch to JSON view, and hit Start.

{
  "banner": "saveonfoods",
  "search_terms": ["eggs", "milk", "chicken breast"],
  "postal_code": "V6M 2P8"
}

Same run over HTTP:

curl -X POST \
  "https://api.apify.com/v2/acts/sunny_eternity~save-on-foods-pricesmart-scraper/run-sync-get-dataset-items?token=YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"banner": "saveonfoods", "search_terms": ["eggs", "milk"], "postal_code": "V6M 2P8"}'

Change banner to pricesmart, urbanfare or qualityfoods to switch chains, or swap search_terms for categories to walk a whole department.

What comes back

{
  "store": "Save-On-Foods",  "name": "Organic Whole Milk 2L",  "price": "5.99",  "unit_price": "$0.30/100ml",  "location_name": "Save-On-Foods Marine Drive",   // named store, not just an id  "selling_type": "by_unit",   // by_unit or by_weight  "comparable_unit_price": 0.3,   // normalized $/100ml  "is_on_sale": false,  … 8 more fields}

All four banners share categories and output fields, since they run on one platform. A dataset from Urban Fare concatenates onto one from Save-On-Foods with no mapping layer.

Four chains, four prices

Same group doesn’t mean same price. The same 3.25% milk can be $7.99 at Save-On-Foods and $8.49 at Urban Fare, same city, same day. Urban Fare sits at the premium end and PriceSmart at the value end, and the gap between them is the thing worth measuring.

Because the input shape is identical across banners, a four-chain sweep is a loop:

from apify_client import ApifyClient

client = ApifyClient("YOUR_API_TOKEN")
rows = []

for banner in ["saveonfoods", "pricesmart", "urbanfare", "qualityfoods"]:
    run = client.actor("sunny_eternity/save-on-foods-pricesmart-scraper").call(run_input={
        "banner": banner,
        "search_terms": ["whole milk 2l"],
        "postal_code": "V6M 2P8",
    })
    rows += list(client.dataset(run["defaultDatasetId"]).iterate_items())

for r in sorted(rows, key=lambda r: float(r["price"])):
    print(r["store"], r["name"], r["price"])

pip install apify-client first. The JavaScript client does the same thing:

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_API_TOKEN' });
const banners = ['saveonfoods', 'pricesmart', 'urbanfare', 'qualityfoods'];

const rows = [];
for (const banner of banners) {
  const run = await client.actor('sunny_eternity/save-on-foods-pricesmart-scraper').call({
    banner,
    search_terms: ['whole milk 2l'],
    postal_code: 'V6M 2P8',
  });
  const { items } = await client.dataset(run.defaultDatasetId).listItems();
  rows.push(...items);
}

Stores, and the rsid

A postal code is usually enough. The Actor reads the chain’s live store directory at runtime rather than a bundled table, so Nanaimo, Kelowna, Prince George, Calgary, Saskatoon and Winnipeg resolve as reliably as downtown Vancouver. Every row names the store and its rsid.

To pin one store, pass locationId. You can read a store’s rsid straight out of the URL on the chain’s own site, the number in /sm/planning/rsid/1982/:

Two things to know

• Coverage stops at four provinces. BC, Alberta, Saskatchewan and Manitoba. A postal code outside the chain’s footprint returns an error naming the provinces that banner actually serves.
• comparable_unit_price is null when a size won’t parse. You get null rather than a guessed number, so filter those rows before sorting on price per 100 ml.

To put these four against the Loblaw banners and T&T in one run instead of four, the comparison API does the matching for you.