Croma Keyword Rank Tracker

Captures the Croma (croma.com, India) search results slate for each keyword, with the rank of every product plus price, MRP, discount, ratings and stock. Use it to see where an electronics brand ranks on the terms shoppers type, and how the competing models around it are priced and reviewed.

What you get

• One row per search result position: keyword, page, position, id, name, brand, price, original_price (the MRP), discount_percent and product_url.
• total_results for the keyword, so share of search has a denominator.
• Rich review data: rating, rating_count and a separate review_count.
• in_stock plus stores_in_stock, the number of Croma stores holding the item.
• Croma's own taxonomy on every row: category_l0, category_l1 and category_l2.
• offer_text for the promotional line on the tile and live_date for when the product went live.

Input

Give the keywords to rank. Everything else has a working default.

{
  "keywords": ["laptop", "smart tv"],
  "locale": "in",
  "pages": 1,
  "delayMs": 1200,
  "callsPerSession": 60,
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["RESIDENTIAL"],
    "apifyProxyCountry": "IN"
  }
}

• keywords (required): search terms to rank. Duplicates are ignored.
• pages: result pages to capture per keyword. Croma serves 21 results a page.
• delayMs and callsPerSession: pacing and session rotation.
• proxyConfiguration: egress for the run. An Indian residential proxy is the default.

Output

One dataset row per search result position.

{
  "platform": "croma",
  "locale": "in",
  "keyword": "laptop",
  "page": 1,
  "position": 1,
  "id": "273404",
  "name": "HP 15 Intel Core i5 13th Gen Thin and Light Laptop",
  "brand": "HP",
  "price": 54990,
  "original_price": 72999,
  "discount_percent": 24,
  "currency": "INR",
  "rating": 4.3,
  "rating_count": 1180,
  "review_count": 214,
  "in_stock": true,
  "stores_in_stock": 37,
  "is_sponsored": false,
  "category_l1": "Laptops",
  "offer_text": "Extra 10% off on HDFC cards",
  "total_results": 1462,
  "product_url": "https://www.croma.com/hp-15-thin-and-light-laptop/p/273404",
  "captured_at": "2026-09-12T03:35:09.757Z"
}

is_sponsored is always false. Croma exposes no sponsored marker on search results, so every position is organic. seller is always null.

Use cases

• Rank tracking for a model or brand across a keyword set, from keyword and position.
• Share of search on category terms, counting your brand against total_results.
• Competitive price benchmarking at rank, pairing position with price and discount_percent.
• Review-strength comparison of the top results from rating, rating_count and review_count.
• Stock depth of the leading results from in_stock and stores_in_stock.

Notes and limits

• Exactly 21 results a page, whatever page size is requested. Croma's service ignores the page-size parameter.
• pages is capped at 10 per keyword, so up to about 210 ranked positions per keyword per run.
• A keyword with no matches is handled cleanly. Croma's search service answers an error rather than an empty list for a term it cannot match, and the actor records zero results for that keyword and carries on.
• Croma exposes no sponsored signal, so is_sponsored is always false by construction. Ad-slot share cannot be measured on this storefront.
• Only the India storefront exists.
• An Indian residential proxy is the default and is recommended. Other egress can be met with an access-denied page.

FAQ

Do I need a proxy? Yes, use an Indian residential proxy. That is the actor default.

How many results can I get? 21 results a page and up to 10 pages per keyword.

Does it need a login or an API key? No. No account and no key.

Can I see which results are ads? No. Croma publishes no paid-placement marker in its search data, so is_sponsored is always false.

What happens to a nonsense keyword? Croma's search service returns an error for a term with no matches. The actor treats that as zero results, logs the keyword and finishes the run normally, so a dead term never fails the whole job.