Amazon Product Scraper (Price, Rating, BSR, Stock) avatar

Amazon Product Scraper (Price, Rating, BSR, Stock)

Pricing

from $4.00 / 1,000 products

Go to Apify Store
Amazon Product Scraper (Price, Rating, BSR, Stock)

Amazon Product Scraper (Price, Rating, BSR, Stock)

Extract Amazon product data by ASIN or URL: price, rating, seller, images and more when pages are accessible. Amazon robot checks can block products; review the ERRORS record and proxy options before larger runs.

Pricing

from $4.00 / 1,000 products

Rating

0.0

(0)

Developer

Murat Uzun

Murat Uzun

Maintained by Community

Actor stats

0

Bookmarked

1

Total users

0

Monthly active users

3 days ago

Last modified

Features

Limited permissions

Share

What is Amazon Product Scraper?

Amazon Product Scraper reads public Amazon product pages by ASIN or product URL and returns one row per product: title, brand, price and list price, rating and rating count, best sellers rank, stock line, seller, images, feature bullets, variants and category. It supports 19 marketplace codes and needs no Amazon login. It is for e-commerce analysts, catalog teams and developers comparing product details. Each accepted product saved to the dataset is one charged row; individual fields can still be missing.

Price: $4 per 1,000 accepted products ($0.004 per saved row). The published README also lists a $0.00005 run start fee; its current billing configuration was not independently verified. See Pricing below for sources and limits.

What data does Amazon Product Scraper extract?

FieldTypeMeaning
asinstringAmazon ASIN
marketplacestringMarketplace code, e.g. com
urlstringProduct page
titlestringProduct title
brandstringBrand from the byline
pricenumberBuy-box price (null when Amazon shows none)
listPricenumberStruck-through list price
currencystringISO code for the marketplace
ratingnumberStars, 0 to 5
ratingCountnumberNumber of ratings
bestSellersRankstringRank text, e.g. #1,234 in Electronics
categoryRanksarrayParsed ranks: {rank, category}
availabilitystringStock line as shown on the page
seller / sellerIdstringSeller name and Amazon seller id
imageUrl / imagesstring / arrayMain and gallery image URLs
featuresarrayFeature bullets
variantsarrayVariant ASINs with labels
isPrimebooleanPrime badge present (see limitations)
categorystringLast breadcrumb

How to use Amazon Product Scraper

  1. Enter ASINs or Amazon product URLs in "Products".
  2. Choose the marketplace (used for ASINs and URLs without a domain).
  3. Run it and download the dataset, or call the Actor through the API.

Example input

{
"products": ["B0DCH8VDXF", "B08JHCVHTY"],
"marketplace": "com"
}

Example output

Real run, 22 Oct 2026, amazon.com, ASIN B0DCH8VDXF (first row of the run; the second ASIN was blocked by Amazon's robot check and was not charged):

[
{
"position": 1,
"asin": "B0DCH8VDXF",
"marketplace": "com",
"url": "https://www.amazon.com/dp/B0DCH8VDXF",
"title": "Apple EarPods Headphones with USB-C Plug, Wired Ear Buds with Built-in Remote to Control Music, Phone Calls, and Volume",
"brand": "Apple",
"price": 17.87,
"listPrice": null,
"currency": "USD",
"rating": 4.5,
"ratingCount": 19258,
"bestSellersRank": null,
"availability": null,
"seller": null,
"variants": null
}
]

Some fields are null in this example: Amazon did not show a stock line, seller or rank on that page for this request. The Actor never estimates a missing value.

Input parameters

  • Products: ASINs or product URLs (…/dp/ASIN).
  • Amazon marketplace: used for ASINs and URLs without a domain.
  • Proxy mode: auto (default) tries direct requests first and switches to a Residential proxy after two robot checks on the same product. never uses direct requests only. always uses Residential from the start. In a 13 Nov 2026 cloud test Amazon answered every direct request with a robot check (6 of 6 ASINs, 0 rows with never); with auto 2 of 3 ASINs came back, using 0.5 MB of Residential traffic in total.
  • Save debug HTML: stores the fetched page for troubleshooting.

Pricing

The Store listing, checked on 13 Nov 2026, displays from $4.00 per 1,000 products. At that rate, 10 accepted rows cost $0.04 in product charges. “Accepted” means a parsed product row saved to the dataset, not a guarantee that every field has a value.

The published README lists a $0.00005 start fee per run. The separate pricing table could not be read during this documentation check, so the currently configured start event and plan-specific totals are unverified. Check the Actor's Pricing tab before starting a run. Platform/proxy usage and product-event charges are different measures; the old $0.017 platform-cost observation is not a customer price estimate. No current total-cost benchmark was measured for this revision.

Amazon Product Scraper vs. alternatives

The Actor figures below are historical observations from the 11 Nov 2026 apify_scorecard supplied for this maintenance review, not a fresh pricing audit. FREE is the Apify plan name, not a claim that these Actors run at no charge. Monthly users (u30) measure usage, not paying customers or output quality.

Option and HTTPS sourceObservation datePlan / price unitu30Comparison scope
WebDataTools Amazon Product Scraper11 Nov 2026FREE: USD 4 / 1,000 products0This Actor; accepted saved product rows, with potentially missing fields.
junglee/amazon-crawler11 Nov 2026FREE: USD 5 / 1,000 results3,617Related Amazon product Actor; result equivalence has not been benchmarked.
delicious_zebu/amazon-product-details-scraper11 Nov 2026FREE: USD 2.50 / 1,000 results298Related Amazon product-details Actor; result equivalence has not been benchmarked.
curious_coder/amazon-scraper11 Nov 2026FREE: USD 1 / 1,000 results238Related Amazon products Actor; result equivalence has not been benchmarked.
Scrapy — BSD license14 Nov 2026Self-hosted open-source software: no software license fee; no hosted per-result planNot measuredDIY alternative: write and maintain your own Amazon spider. Hosting, proxy traffic and engineering may cost money; Amazon success and field coverage were not tested here.

On 14 Nov 2026, the linked Store headers displayed from USD 3, USD 1.50 and USD 0.75 per 1,000 results for Junglee, Zebu and Curious Coder respectively. Those headers did not establish the FREE-plan rate, so they do not replace the dated plan-specific observations above. Check each Actor's current Pricing tab before choosing.

A competitor's “result” has not been shown to equal our accepted “product” row. No matched-input quality, total-cost or 2x advantage benchmark was performed. Compare field completeness, blocked requests, missing prices, start fees and any platform/proxy costs for your own workload; these listed units alone do not establish price or quality superiority.

JavaScript API example

With Node.js 18+ and APIFY_TOKEN supplied through your environment, save this as example.mjs and run node example.mjs:

const input = {
products: ['B0DCH8VDXF'],
marketplace: 'com',
proxyMode: 'auto',
};
const response = await fetch(
'https://api.apify.com/v2/acts/webdatatools~amazon-product-scraper/run-sync-get-dataset-items',
{
method: 'POST',
headers: {
Authorization: `Bearer ${process.env.APIFY_TOKEN}`,
'Content-Type': 'application/json',
},
body: JSON.stringify(input),
},
);
if (!response.ok) throw new Error(`Apify HTTP ${response.status}`);
console.log(await response.json());

Python API example

Python 3, standard library only; set APIFY_TOKEN in your environment before running:

import json
import os
import urllib.request
request = urllib.request.Request(
"https://api.apify.com/v2/acts/webdatatools~amazon-product-scraper/run-sync-get-dataset-items",
data=json.dumps({
"products": ["B0DCH8VDXF"],
"marketplace": "com",
"proxyMode": "auto",
}).encode("utf-8"),
headers={
"Authorization": "Bearer " + os.environ["APIFY_TOKEN"],
"Content-Type": "application/json",
},
method="POST",
)
with urllib.request.urlopen(request, timeout=310) as response:
print(json.load(response))

CLI example (curl)

With curl installed and APIFY_TOKEN set, this command starts one run and returns dataset rows:

curl --fail-with-body --max-time 310 \
--request POST \
'https://api.apify.com/v2/acts/webdatatools~amazon-product-scraper/run-sync-get-dataset-items' \
--header "Authorization: Bearer ${APIFY_TOKEN}" \
--header 'Content-Type: application/json' \
--data '{"products":["B0DCH8VDXF"],"marketplace":"com","proxyMode":"auto"}'

These are alternative examples: each execution starts a separate billable run. The synchronous API waits up to 300 seconds. On timeout or connection loss, inspect the existing run in Console before retrying. For longer jobs, use the asynchronous runs API and retrieve the completed run's dataset. An empty dataset does not explain the failure by itself: inspect OUTPUT and ERRORS in the run's key-value store.

MCP example for AI agents

Connect your MCP client to https://mcp.apify.com and authorize with Apify. Then use this concrete prompt:

Inspect webdatatools/amazon-product-scraper with fetch-actor-details. Run it once with products ["B0DCH8VDXF"], marketplace "com", and proxyMode "auto" using call-actor. Wait for that run to finish; use get-actor-run if it is still running. Read its dataset with get-dataset-items and show asin, title, price and currency. If no rows were saved, read OUTPUT and ERRORS with get-key-value-store-record. Do not start a second run automatically.

As documented in the Apify MCP guide, call-actor returns status and storage IDs; dataset retrieval is a separate step. These API/MCP examples have not been executed against the live build in this revision.

Errors and the OUTPUT record

Products rejected before a dataset write incur no product-row charge. If a dataset write acknowledgement times out, delivery and charge are unknown. They are listed in the ERRORS record, and every run writes an OUTPUT record with products, requested, withPrice, failed, proxyMode, errors and a reason when nothing was saved. If all supplied products are invalid, the run fails with INVALID_INPUT after writing these diagnostic records. Mixed valid and invalid products can still succeed with partial results. Other runs without rows still succeed, unless the run deadline or a finalization failure prevents completion; their status message says why.

errorCodeMeaningWhat to do
INVALID_INPUTNot an ASIN or an Amazon product URLUse a 10-character ASIN or a …/dp/ASIN link
BLOCKEDAmazon showed a robot check on every sessionKeep Proxy mode auto (or always) and run again later
DEADLINEProcessing stopped before the platform timeoutInspect saved rows and diagnostics before retrying unfinished products
NOT_FOUNDNo product page: not found, or not sold in this marketplaceCheck the ASIN and the marketplace

The Actor reads and logs the platform's timeoutAt and reserves the last 10 seconds for shutdown. It stops requests/retry waits before that window and bounds each diagnostic/status operation to at most 2 seconds (and the remaining platform time). A deadline with zero saved rows fails; saved rows are retained and the terminal message says Partial (DEADLINE). OUTPUT.reason explains the cutoff. If a storage write does not acknowledge in time, finalization fails rather than claiming successful delivery; inspect the dataset before retrying. Diagnostic persistence cannot be guaranteed if storage is unavailable or the platform has already killed the process. When the platform supplies no valid end time, the Actor derives and logs timeoutAt from .actor/actor.json → defaultRunOptions.timeoutSecs (currently 3600 seconds), added to the platform start time or, if absent, the process startup time captured before SDK initialization. This fallback is the configured default, not a claim to know an unavailable platform override. A missing or non-positive fallback duration fails instead of silently removing the run limit.

FAQ

What do 1,000 accepted products cost? $4 in product charges at the displayed rate. This is not a verified all-inclusive invoice; check the current Pricing tab for the start fee and plan terms.

Do blocked or invalid products incur a product charge? They produce no dataset row and therefore no product-row charge. A run can still have a start fee or other applicable charges. All-invalid input fails; a mix of valid and invalid entries can return partial results.

Is a product with a null price charged? Yes, if its product row was accepted and saved. Billing is per product row, not per populated field.

How can I control cost? Start with one ASIN and set the platform's maximum cost per run in Console. proxyMode: "never" avoids Residential retries but can return no products. auto may use Residential traffic after failed direct attempts; actual total cost is not benchmarked for this revision.

Does it need an Amazon account? No.

Why is the price null? Amazon shows no buy-box price for some products in some regions or at some times (for example "Currently unavailable"). The Actor returns null instead of a guessed price.

Why are some ASINs failing? Amazon sometimes answers with a robot check. Each product gets several fresh sessions (homepage visit first). In the 23 Oct 2026 cloud run 1 of 20 ASINs (B08JHCVHTY) got a robot check on every session; it is listed in the ERRORS record and was not charged. Retry it later.

Limitations. Verified in the cloud on 23 Oct 2026: amazon.com, 20 ASINs taken from the Best Sellers Electronics page, 19 rows returned. Field coverage over those 19 rows: title 19, price 19, rating 19, list price 10, seller 11, availability 4, Best Sellers Rank 0 (the pages served to the cloud carried no rank text). isPrime is unverified. Other marketplaces were not run in the cloud.

  • Amazon Search Scraper (keyword results with sponsored flag)
  • Amazon Best Sellers Scraper (category rank lists)

Support

Open an issue on the Actor page with the ASIN and marketplace so a bad result can be reproduced.

Please leave an honest review on the Apify Store page about your experience; no reward or incentive is offered.