Google Shopping price tracking

1. 1. Choose queries and a country

   Use the searches your shoppers type: a category, such as wireless earbuds, or a product name. country sets where the search is made from, which sets the stores and the currency. Keep both fixed between runs so changes come from Google, not your settings.

   GET /v1/google-shopping?q=wireless+earbuds · query

   {
     "q": "wireless earbuds",
     "language": "en",
     "country": "us"
   }

2. 2. Record each product card

   One request returns the products Google shows for the query, usually 20 to 30. Each card has product_id, Google's id for the product across every store, plus the headline store in merchant, price, original_price and badge for a discount, delivery, and rating. When a card is priced from several stores, merchant is null and price is the lowest of them.

   GET /v1/google-shopping?q=wireless+earbuds · results.0

   {
     "position": 1,
     "title": "JLab GO Air Pop True Wireless Earbuds",
     "product_id": "9146411353573113826",
     "price_text": "$14.99",
     "price": 14.99,
     "currency": "USD",
     "original_price_text": "$20",
     "original_price": 20,
     "badge": "24% OFF",
     "merchant": "Target",
     "more_merchants": true,
     "delivery": "Free delivery on $35+",
     "rating": 4.3,
     "ratings_count": 33000,
     "section": "Popular products"
   }

3. 3. Compare day to day by product_id

   product_id is stable between searches, so join today's snapshot to yesterday's on it. A changed merchant means another store now leads the card; a new badge or original_price marks a discount; a product that drops out of the results has lost its place for that query.

   GET /v1/google-shopping?q=wireless+earbuds · results.1

   {
     "position": 2,
     "title": "Beats Solo Buds Wireless Earbuds",
     "product_id": "549048356659243400",
     "price_text": "$59.99",
     "price": 59.99,
     "currency": "USD",
     "original_price_text": "$80",
     "original_price": 80,
     "badge": "25% OFF",
     "merchant": "Best Buy",
     "more_merchants": true,
     "delivery": "Free delivery by Thu",
     "rating": 3.8,
     "ratings_count": 6200,
     "section": "Popular products"
   }

# Daily Google Shopping snapshot for a set of product queries: which store
# heads each product card, at what price and discount.
import csv, datetime, os, requests

API = "https://api.clair.im"
HEADERS = {"Authorization": f"Bearer {os.environ['CLAIR_API_KEY']}"}
QUERIES = ["wireless earbuds", "air fryer", "robot vacuum"]
COUNTRY = "us"  # sets the stores and the currency

def shopping(q):
    r = requests.get(f"{API}/v1/google-shopping", headers=HEADERS, timeout=90,
                     params={"q": q, "country": COUNTRY})
    r.raise_for_status()
    return r.json()["results"]

today = datetime.date.today().isoformat()
with open("shopping.csv", "a", newline="") as f:
    out = csv.writer(f)
    for q in QUERIES:
        for card in shopping(q):
            out.writerow([today, q, card["position"], card["product_id"],
                          card["title"], card["price"], card["currency"],
                          card["original_price"], card["badge"],
                          card["merchant"], card["more_merchants"],
                          card["rating"], card["ratings_count"]])

Set CLAIR_API_KEY to your Clair key; one plan covers the Google Shopping API. Each call spends that API's rate from the plan's monthly balance.

• Each card shows one store's offer. The full list of stores selling a product, with each price, is not returned.
• Google shows one page of products per query, so there is no second page to read. Narrow the query to see more of a category.
• Location is set per country. There is no city or postcode targeting, and delivery terms are those Google shows for the country.
• ratings_count is rounded as Google shows it: 6.2K is 6,200.
• There are no product images: Google embeds the thumbnails in the page rather than linking them.
• Google Shopping lists products from stores registered with Google Merchant Center. A product missing from the results may still be sold by stores that are not.