When Google shows a shopper a product, it does not show a link list anymore. It opens an immersive detail panel: price range across stores, star-by-star rating breakdown, specs, variants, videos, and sample reviews, all on one screen. That panel on Google Shopping is the best free product research page on the internet, and there is no endpoint for it. The Google Immersive Product API on Apify fixes that: send a product query, get those rich pages back as structured JSON.
Disclosure: the Apify links in this post are affiliate links. If you run the Actor, I may earn a referral commission at no extra cost to you.
Does Google have a product data API?
Only for your own catalog. The Merchant and Content APIs exist so sellers can push listings into Google, not so developers can read the shopper-facing product panels out. There is no official way to request "the immersive page for wireless earbuds" and get prices, ratings, and specs across sellers. So a Google product AI data feed in practice means a scraper you call like an API: query in, product records out.
What the Google Immersive Product API returns
The Google Immersive Product API returns one record per matching product as structured JSON: title, brand, price range across stores, aggregated rating with a star-by-star breakdown, specs, variants, seller listings, related videos, and sample user reviews.
| Field | Example | Notes |
|---|---|---|
| title | "Sony WH-1000XM5" | Product name as Google shows it |
| brand | "Sony" | Manufacturer |
| price range | $278 to $399 | Across the stores selling it |
| rating | 4.6, with 5-star to 1-star counts | Aggregated, with review counts |
| specs | driver size, battery life, weight | Product specifications |
| stores | seller listings | Who sells it and where |
Sample user reviews, variant information, and thumbnails ride along in the same record.
Who this is for
E-commerce teams doing competitive intelligence on price and rating position, analysts tracking a product's review momentum over time, and developers feeding an AI shopping agent that needs real specs and prices instead of guesses.
The manual way, and where it breaks
By hand, you search the product, open the panel, and copy what you need, which is fine exactly once. Automating it is the trap: the immersive panel renders through layers of JavaScript, the markup is deeply nested and changes without notice, and the interesting parts (the star breakdown, the seller list) load on interaction. My first attempt returned beautiful HTML with none of the data in it. Add proxies and retries, and the side project becomes an infrastructure project.
The faster way: run the Google Immersive Product API
Apify Console
- Open the Google Immersive Product API and click Try for free.
- Enter a
querylike "wireless earbuds", or a batch inqueries. - Run it and download the dataset as JSON, CSV, or Excel.
REST
curl -X POST "https://api.apify.com/v2/acts/johnvc~google-immersive-product-api/runs?token=YOUR_APIFY_TOKEN" \
-H "Content-Type: application/json" \
-d '{ "query": "wireless earbuds", "maxResultsPerQuery": 5 }'
Run endpoint reference: the Apify API docs. The task Get rich product detail pages from a Google search is this exact call saved and runnable.
Get product pages in Python
import json
from apify_client import ApifyClient
client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("johnvc/google-immersive-product-api").call(
run_input={
"queries": ["apple watch", "wireless earbuds"],
"gl": "us",
"maxResultsPerQuery": 5,
}
)
for product in client.dataset(run["defaultDatasetId"]).iterate_items():
print(json.dumps(product, indent=2)[:400])
print("---")
Each record is one product's full immersive page: title, brand, price range, the rating breakdown, specs, and the stores carrying it.
Get a product's ratings and reviews from Google
Ratings with the star-by-star breakdown are the fastest read on how a product is actually doing. The task Get a product's ratings and reviews from Google pulls exactly that slice.
Price and rating snapshots by category
The same recipe works across categories, and three saved tasks show it: Apple Watch prices and ratings as JSON, headphone prices and ratings as JSON, and laptop prices, brands, and ratings as JSON. Swap the query and you have your own category tracker.
Use it from Claude and other MCP clients
The Actor is MCP-ready, so Claude, Claude Code, and Cursor can call it as a live tool: ask "compare the top five wireless earbuds by rating and price range" and the agent pulls current product pages instead of reciting training data. You can read more about Claude and Claude Code at claude.ai.
FAQ about scraping Google product pages
Is there a free Google product scraper?
Free credit gets you started: the Actor bills per product returned, and new Apify accounts include platform credit that covers early runs. maxResultsPerQuery caps how many products a query returns (default 10, max 20), so spend is bounded before a run starts.
How does the scraper get product reviews and ratings?
Each product record includes the aggregated rating, the star-by-star breakdown with review counts, and sample user reviews from the immersive panel. That is usually enough to spot a review-bomb or a quality slide without scraping individual review pages.
Can this scraper feed an AI shopping agent?
Yes, that is one of the design goals. One call returns the full product detail record, so an agent gets title, price range, specs, and rating in a single tool response instead of five browsing steps.
Does the scraper work in Claude via MCP?
Yes. Connect the Apify MCP server and the Actor appears as a callable tool in Claude, Claude Code, or Cursor, ready for live product lookups mid-conversation.
Can I schedule the scraper to monitor prices and ratings?
Yes. Save a query set as a task, attach an Apify schedule, and each run appends fresh records, which gives you price-range and rating history for the products you track. Start from the Google Immersive Product API.
What are the limits of this product scraper?
It returns the top matching products for a query, up to 20 per query, so it is a research and monitoring tool rather than a full catalog crawler. Prices come back as ranges across the stores Google lists at lookup time, and what you see is Google's view of the product, localized by your gl and hl settings.
More from Truffle Pig Data
Two sibling Actors cover the flat side of Google's shopping results: the Google Shopping API for products, prices, and deals from shopping searches, and the Google Shopping Lite API for a cheaper, lighter version of the same lookup.
Wrapping up
Google already built the richest product page in e-commerce; the missing piece was a way to query it. The Google Immersive Product API returns it as JSON, one record per product.
Top comments (0)