How do I scrape Amazon product data without getting blocked?

Use residential or ISP-grade IPs, throttle per-domain to a few requests per IP per minute, and rotate browser fingerprints between requests. Amazon blocks datacenter IPs quickly and serves CAPTCHAs to repeat identities. A managed API like Ollagraph handles residential egress, fingerprint rotation, and JS rendering by default, so you send a URL and get clean JSON back.

What is Amazon SERP data?

Amazon SERP data is the structured result of an Amazon search — the ranked list of products for a keyword, with ASIN, title, price, rating, review count, sponsored flag, and position. Teams use it to track share of search, monitor competitors' organic and sponsored placement, and measure keyword rank over time. You collect it by fetching the search results page and extracting each result card.

Can I extract Amazon product fields without writing a parser?

Yes. Point the structured-extract endpoint at a product URL and it returns title, price, currency, availability, rating, and review count as JSON — it reads JSON-LD, Open Graph, and microdata the page already exposes, so you don't maintain brittle CSS selectors that break on Amazon's frequent layout changes.

Is scraping Amazon legal?

Observing publicly displayed product information is generally lawful in the United States and most jurisdictions — the data is shown to any visitor. Redistribution, or use that violates a specific contract, raises separate questions. Most teams keep the data internal to pricing, merchandising, and market-research decisions and stay well inside the law. Talk to counsel before commercial redistribution.

How do I collect prices for thousands of ASINs on a schedule?

Use the batch endpoint (up to 100 URLs per call) for the fetch, then fan out per-URL structured-extract calls from a worker pool, and move to the async endpoint with webhook delivery once the pipeline is real. Schedule the run from your own cron. Billing is one credit per call with failed calls auto-refunded, so a large ASIN set stays predictable.

Do I need a headless browser to scrape Amazon?

Sometimes. Static product fields are often available without rendering, but search results, some price widgets, and buy-box variations are JavaScript-rendered. Ollagraph auto-detects when a page needs rendering and escalates transparently — the same request works whether or not the page needs a browser, and the response tells you which path served it.

How to scrape Amazon product data in 2026

A working playbook for pulling Amazon product fields and search-results (SERP) data at scale — price, rating, reviews, ASINs, and rank — using structured extraction and a managed scraping API, without building a proxy farm.

What scraping Amazon product data looks like in 2026

Scraping Amazon in 2026 is a managed pipeline, not a parser you babysit. You send a product URL or ASIN to a structured-extract endpoint and get back price, availability, rating, and review count as clean JSON; you send a search URL and get back the ranked result cards. The residential IPs, fingerprint rotation, and JavaScript rendering that used to be the hard part are bought, not built, by most teams operating below eight-figure monthly request volumes.

That is the short answer. Below is the operating manual for the pricing, merchandising, and market-research teams who have to make it real.

The problem behind the problem

Amazon is the single richest source of e-commerce demand signal on the web — and one of the most aggressively defended. Teams want three things from it: product fields (price, availability, rating, review velocity) for a known set of ASINs, and search-results data (which products rank for which keywords, organic and sponsored) to measure share of search. The blocker is never the parsing logic; it is staying unblocked at scale while Amazon rotates layouts, serves CAPTCHAs, and fingerprints datacenter traffic.

The homegrown version of this quietly becomes a proxy-management project with a scraping feature attached. The managed version is a URL in, JSON out.

What the leading alternatives do well

This is a mature category and several vendors are genuinely good. If you're evaluating buy versus build, your shortlist likely includes some of these.

Rainforest API

Rainforest is an Amazon specialist and it shows — dedicated endpoints for product, offers, reviews, and search, with response shapes designed around Amazon's own structure. If Amazon is the only site you need and you want the most turnkey Amazon-shaped output available, they are a strong, focused choice.

Oxylabs / broad scraping platforms

The larger scraping platforms bundle e-commerce-specific parsers with big proxy networks and enterprise support. For organizations that need Amazon alongside many other retail sites under one procurement relationship, that breadth and the account management that comes with it are real advantages.

Build it yourself

Renting a residential proxy pool, running a headless-browser fleet, and writing the parsers gives you total control — and the largest hidden bill in the category once you load in proxy spend, compute, and the engineering attention it quietly consumes every time Amazon ships a layout change. Most teams underestimate that last line.

Where Ollagraph goes further

The Amazon-specialist tools are excellent if Amazon is your whole world. Ollagraph wins for the teams whose pipeline spans Amazon and Shopify stores, direct-to-consumer sites, and Google Maps listings — one API, one key, one flat per-call price, feeding your warehouse or webhook rather than ending at someone else's dashboard.

Three concrete outcomes. First, coverage you control — you choose the URLs, the cadence, and the fields, and the generic structured extractor handles product pages across platforms without a per-site parser. Second, predictable economics — one credit per call, failed calls auto-refunded, no per-ASIN tier or surprise overage. Third, no proxy project — residential egress, fingerprint rotation, and JS rendering are included, so a datacenter block is never your problem.

The recipe, step by step

Step 1. Provision an API key

Sign up on the pricing page, copy your key from the dashboard, and store it as an environment variable. New accounts start with 1,000 free credits, no card.

export OLLAGRAPH_API_KEY="osk_xxxxxxxxxxxx"

Step 2. Extract fields from a single product

Start with one ASIN to confirm the pipeline. The structured-extract endpoint returns clean fields without a parser — it reads the JSON-LD, Open Graph, and microdata the product page already publishes.

curl -X POST https://api.ollagraph.com/v1/extract/structured \
  -H "Authorization: Bearer $OLLAGRAPH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://www.amazon.com/dp/B08N5WRWNW"
  }'

{
  "title": "Echo Dot (4th Gen)",
  "price": 49.99,
  "currency": "USD",
  "availability": "In Stock",
  "rating": 4.7,
  "review_count": 318241
}

Step 3. Pull Amazon search-results (SERP) data

To track share of search and keyword rank, fetch the search-results page. Search is JavaScript-rendered, so let the managed engine render it and return the HTML; then extract each result card (ASIN, title, price, rating, sponsored flag, position). The same request works whether or not rendering is needed — the response's rendered_via field tells you which path served it.

curl -X POST https://api.ollagraph.com/v1/scrape \
  -H "Authorization: Bearer $OLLAGRAPH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://www.amazon.com/s?k=wireless+earbuds",
    "render": "auto"
  }'

Parse the result cards for organic versus sponsored placement and store position by keyword and date. That time series is your Amazon SERP dataset — the basis for share-of-search and rank-tracking reports.

Step 4. Run the full ASIN set in batch

For a scheduled pull, use the batch endpoint — up to 100 URLs per call. For 1,000 ASINs that's ten calls per run.

curl -X POST https://api.ollagraph.com/v1/scrape/batch \
  -H "Authorization: Bearer $OLLAGRAPH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "urls": [
      "https://www.amazon.com/dp/B08N5WRWNW",
      "https://www.amazon.com/dp/B09B8V1LZ3",
      "https://www.amazon.com/dp/B0BDHWDR12"
    ],
    "format": "json",
    "concurrency": 10
  }'

Step 5. Switch to async with webhook delivery

Once the pipeline is real, stop polling synchronously. Submit jobs to the async endpoint and let results arrive at a webhook you control the moment they're ready. Schedule the run from your own cron.

curl -X POST https://api.ollagraph.com/v1/scrape/async \
  -H "Authorization: Bearer $OLLAGRAPH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://www.amazon.com/dp/B08N5WRWNW",
    "webhook_url": "https://your-app.example.com/hooks/amazon"
  }'

Verify the payload signature, write the record to your warehouse, and diff against the last stored price and review count. A jump in review velocity is often an earlier demand signal than the price itself.

The pricing math

Ten thousand ASINs pulled hourly is 7.2 million requests a month; most teams don't need hourly for product fields — a few times a day is plenty, which brings it well down. Billing is one credit per call with failed calls auto-refunded, so the bill scales linearly with what you actually pull and there is no per-ASIN tier. Compare line by line on the pricing page.

What can go wrong

Sale price versus list price are sometimes both on the page — decide which one your decision needs and extract it consistently. Buy-box price can differ from the headline price when third-party sellers win the box, so capture both if your use case is competitive. Out-of-stock is noisy — confirm across a couple of polls before treating a stockout as an event. And search results are personalized and localized, so pin a country and keep the request shape stable to keep your rank time series comparable.

For the legal posture around observing public product pages, the US case law on public-web access — summarized accessibly at en.wikipedia.org/wiki/HiQ_Labs_v._LinkedIn — is the foundational reading. Talk to counsel before commercial deployment; the general direction is favorable for public-page observation.

Pairing Amazon data with the rest of your pipeline

Amazon is one feed in a larger commercial-intelligence layer. Teams commonly pair it with competitor pricing across other storefronts — see the competitor-pricing recipe — and with catalog or review feeds via the generic extract endpoint. Pricing alone tells you what competitors do; pricing plus review velocity tells you whether it's working.

What to do next

Pick ten ASINs and one search keyword. Sign up, paste the curl from Step 2, and confirm real fields come back in the next ten minutes. Then add search-results collection in week one, batch your full ASIN set in week two, and wire the webhook once the feed is load-bearing. Read the docs and explore the actor catalog for the platforms already packaged as one-call endpoints.