Amazon reviews API: what Amazon allows, and how to get review text

A competitor's one-star reviews give you free ideas for improving your product. Buyers tell you the lid breaks, the battery dies after a month, or the tub is smaller for the same price. If you research products for an agency, sell software to Amazon brands, or track complaints on your listings, you need those reviews in a spreadsheet or database instead of 40 browser tabs.

Amazon's APIs serve affiliates and sellers managing their accounts. None returns review text. I checked their limits in Amazon's docs, ran 25 calls through AnyAPI's Amazon endpoints on six products, and checked the other services' pricing pages.

Each option suits a different job.

The best Amazon review APIs

• Amazon Creators API for affiliate sites
• Amazon Selling Partner API for sellers who want review topics
• AnyAPI for review text by ASIN, billed per call
• Bright Data for bulk review records
• Apify Amazon review scrapers for star and keyword filters on a schedule
• SerpApi for Amazon's review summary next to product data

What makes a good Amazon reviews API?

If you need more reviews than you can copy by hand, these APIs can help. You might be researching competitors, checking complaints before designing a product, tracking your listings, or analyzing reviews for clients. I compared these features:

• Review text, not only the star average. The average tells you a product is 4.1 stars. The text tells you why.
• How many reviews you get. Amazon now requires sign-in for full review pages. Signed-out visitors see only the few reviews on the product page. Traject Data's Rainforest API dropped one of its review endpoints because Amazon's newest reviews are no longer public. Check how many a vendor returns before relying on a claim of "up to 500".
• Filters that work. Sort by newest, filter for one to two stars, or search for "battery" to find complaints without reading everything.
• Country sites. A brand selling in the US, UK, and Germany needs all three.
• Price per review, not per call. Getting two reviews for ten times the price is much worse value than getting ten.

The best Amazon review APIs at a glance

Best Amazon reviews API for affiliate sites

Amazon Creators API

Amazon Creators API pros:

• Amazon's own product data, with no page to read
• Free for Associates
• Product details, offers, and categories for the products you link to

Amazon Creators API cons:

• No reviews at all, not even star ratings
• Needs an Associates account and 10 qualifying sales in the last 30 days

Affiliate sites used the Product Advertising API for years. Amazon has deprecated it and directs Associates to its replacement, the Creators API.

Use this API if you run a deals site, review blog, or gift guide and already earn Amazon commissions. You can search the catalog or look up ASINs for prices, images, and features to show beside affiliate links.

It won't give you what customers wrote. You can search items, get details, list variations, and browse categories, but you can't get reviews or star ratings to analyze complaints or track buyer feedback.

Amazon requires an Associates account and 10 qualifying sales in the past 30 days.

Amazon Creators API pricing: no charge. Access needs an Associates account with qualifying sales. Source: Amazon Creators API docs.

Best Amazon reviews API for sellers who want review topics

Amazon Selling Partner API

Amazon Selling Partner API pros:

• Amazon's review analysis: the ten most positive and ten most negative topics per product
• Weekly trends in what customers praise and complain about, and why they return things
• A separate API to ask buyers for a review after delivery

Amazon Selling Partner API cons:

• Needs a seller or vendor account with the Brand Analytics or Selling Partner Insights role
• Topics with up to three short snippets each, never the full reviews
• English only, in seven marketplaces

Sellers and their software use the Selling Partner API to manage orders, listings, inventory, ads, and reports. Two parts handle reviews.

The Customer Feedback API analyzes reviews and returns, using the same data as Amazon's Product Opportunity Explorer for sellers. It returns a product's ten most positive and ten most negative review topics, with up to three review snippets per topic. Data refreshes weekly and covers the US, UK, France, Italy, Germany, Spain, and Japan.

If you sell on Amazon or have access to a client's account, try this first. Amazon groups the complaints for free, which may answer your question without full review text.

The Solicitations API sends Amazon's review request to a buyer after delivery. You use it to request reviews, not read them.

It never returns full reviews. Amazon's topics and up to three snippets each help you spot themes, but you can't read buyers' full accounts. You also need a seller or vendor account with the right role.

Amazon Selling Partner API pricing: no charge. The Customer Feedback API needs the Brand Analytics role for vendors or the Selling Partner Insights role for sellers. Source: Amazon SP-API docs.

Best Amazon reviews API for review text by ASIN, billed per call

AnyAPI

AnyAPI pros:

• Review text, star rating, date, helpful votes, and the verified-purchase flag, by ASIN or product link
• Supports 22 Amazon country sites, including the US, UK, Germany, Japan, India, and Brazil (I tried the US and UK)
• Product details and a seller's storefront on the same key, billed per call from a prepaid wallet

AnyAPI cons:

• About ten reviews per call, from the handful Amazon shows signed out
• Star, keyword, or date filters use a more expensive source
• No page two to get the next ten reviews

Use AnyAPI to track reviews across many products for clients or software you sell. Send an ASIN or product link to amazon.reviews for the same JSON format across products and country sites.

curl -sS -X POST 'https://api.getanyapi.com/v1/run/amazon.reviews' \
  -H "Authorization: Bearer $ANYAPI_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{"product":"B0D1CXL52G","limit":10}'

This response includes one shortened review of Ninja's Flip toaster oven and the surrounding fields:

{
  "output": {
    "found": true,
    "data": {
      "items": [
        {
          "rating": 3,
          "title": "Not a great toaster.",
          "text": "I wanted to love this. As a toaster it's not great. One side is good, the other side not so much. The air fryer works well. Great for reheating pizza. I love it can be flipped to get out of the way. Not easy to clean but I read that from reviews.",
          "verifiedPurchase": true,
          "helpfulVotes": 3,
          "createdUtc": 1793059200
        }
      ]
    }
  },
  "costUsd": 0.0013,
  "items": 10
}

Each call returns about ten reviews, whatever limit you set. All six products returned ten when called without a limit, which requests up to 50. Eight matched the ten-review call. A call without filters costs the same base price at any limit, so leaving out limit costs nothing extra.

Sorting newest first at the same price returned nine of the same ten reviews in a different order, plus one new review. Filtering for one and two stars found two reviews for JBL's Vibe Beam 2 earbuds, which have more than 8,000 ratings, and three for the Ninja toaster oven. Those calls cost 11 to 14 times as much as a plain call. Searching for "battery" cost about 18 times as much and returned four reviews.

Filters find reviews the plain call misses. The JBL star and keyword calls added six reviews to the plain ten. Across four calls, I collected 18 different reviews, still few compared with 8,000 ratings.

Plain calls can leave out limit. If you sort, pick a country site with region, or ask for verified purchases only, keep limit at 10 or below; above 10, those calls use a more expensive source. Add a star or keyword filter for specific reviews, such as those mentioning "battery".

The ten reviews include a mix of ratings. Across the six products, 41 of 60 reviews were five stars, and five of the six products had a one-star review in their ten.

The complaints gave you specific problems to address. An Isopure buyer said the tub shrank from 3 pounds to 2.5 while the price rose. An Oral-B owner said it cleans well but is "not fun to clean afterwards". A five-star JBL reviewer found the bass weaker than expected. You'll miss that complaint if you sort by star rating alone.

To build a longer history, save each review and repeat calls on a schedule. Every product's ten included reviews from the past few weeks, so new ones may appear over time. I only tested on one day.

A UK product link returned six reviews, including Italian, Spanish, and Arabic reviews from other countries shown on the UK page. US review calls took about 6 to 15 seconds each; filtered calls took 40 to 50.

The storefront call, amazon.seller_products, lists a seller's products, ratings, and review counts so you can quickly choose which products to collect reviews for.

AnyAPI pricing: $1.30 per 1,000 calls of up to ten reviews, or $0.13 per 1,000 reviews when each call returns ten. Calls the cheapest source can't answer use a more expensive source, up to $0.226 per call. This includes star, keyword, or date filters. Product details cost $0.80 per 1,000 and seller storefronts $0.00085 a page. You prepay without a plan. New accounts get free credit without a card.

Best Amazon reviews API for bulk review records

Bright Data

Bright Data pros:

• 5,000 free records a month, no card
• Pay as you go at $1.50 per 1,000 records, where one record is one review
• A max_reviews input for how many reviews to collect per product

Bright Data cons:

• No star, sort, or keyword filter in its inputs
• No word on whether Amazon's sign-in limits how many reviews it gets

Bright Data sells an Amazon reviews scraper. Send product links, set a review count per product, and pay for each returned review record.

If you need a large weekly review file for a category, start with the free 5,000 records a month. Check how many reviews it returns for a familiar product before relying on it. Its pages don't say whether Amazon's sign-in limits collection.

It also sells more than 43 million Amazon reviews as a dataset, with a $250 minimum order, for historical analysis rather than tracking new reviews.

Bright Data pricing: 5,000 free records a month; then $1.50 per 1,000 records pay as you go, or $499 a month for 384,000 records with extra records at $1.30 per 1,000. Source: Bright Data's Amazon reviews scraper page.

Best Amazon reviews API for star and keyword filters on a schedule

Apify Amazon review scrapers

Apify Amazon review scrapers pros:

• Several actors to pick from, most with star, keyword, and sort filters
• Runs on Apify's scheduler and writes to a dataset you can export
• Apify's free plan includes $5 of usage a month

Apify Amazon review scrapers cons:

• Each actor is a different author's code, with its own fields, limits, and price
• They claim hundreds of reviews per product but don't explain how they get past Amazon's sign-in requirement

An Apify actor is a scraper published to its store. You supply inputs, start a run, and get a dataset. Junglee's Amazon Reviews Scraper is the most-used Amazon reviews actor, with about 2,400 monthly users, followed by Axesso's.

Both claim more than ten reviews. Junglee's claims up to 100 per star rating and 500 per product, with star, keyword, and recent-or-helpful filters. Axesso's stops at ten pages, citing Amazon's limit. I didn't test either, so these are the authors' claims. Try a familiar product before building on one.

If your team already uses Apify, these suit weekly exports to a spreadsheet or warehouse. Check each actor's pricing tab; prices vary several times over.

Apify Amazon review scrapers pricing: per review, by actor and plan. Junglee's is $6 per 1,000 reviews on the free plan, down to $1 on the top plan, with an extra charge for date filters; Axesso's is $0.90 per 1,000. Apify plans start free with $5 of usage. Source: each actor's Apify Store pricing tab.

Best Amazon reviews API for Amazon's review summary next to product data

SerpApi

SerpApi pros:

• Amazon's own AI summary of the reviews, with themes and snippets, in a product lookup
• The top reviews from the product page in the same response
• One key for Amazon, Google, and other search engines

SerpApi cons:

• No reviews endpoint of its own; reviews come as part of a product lookup
• Billed by monthly plan, and about $25 per 1,000 lookups on the smallest one

SerpApi is a search results API. Its Amazon product engine returns the page's review section: Amazon's "Customers say" summary, review themes with snippets, and top reviews.

If you already pay for SerpApi's Google results, you can easily add Amazon's summary of what buyers like and dislike. If you only need reviews, you're paying monthly for the same top reviews the others return.

SerpApi pricing: 250 free searches a month; Starter is $25 a month for 1,000 searches and Developer $75 for 5,000. Source: serpapi.com/pricing.

Which Amazon reviews API should you use?

• You run an affiliate site and want product data for your links: the Amazon Creators API.
• You sell on Amazon and want Amazon's own summary of review topics: the Selling Partner API's Customer Feedback API.
• You need review text for any product, in many countries, paid by the call: AnyAPI.
• You want a big weekly file of reviews by the record: Bright Data.
• You already use Apify and want star or keyword filters on a schedule: an Apify reviews actor.
• You already pay for SerpApi and want Amazon's review summary: SerpApi's product engine.

Save reviews as you go; no single call returns a product's full history. The Amazon API guide covers search, product details, and ASIN lookups. The Amazon reviews API page lists every input.

Use data responsibly and follow AnyAPI's Acceptable Use Policy.