Amazon Search Scraper API: Request, Response, and Credit Planning
Amazon search-result pages are a compact view of a market: which products rank, at what price, with what rating and badge, in which country.
Collecting that view at scale is the part most teams underestimate. A managed scraping API exists to take the browser, proxy, and anti-bot work off your plate so the pipeline can focus on the data itself. This guide explains what such an API actually accepts and returns, how its credit model stacks, and where the failure modes sit, so you can decide whether it fits your workflow before you write a line of code.
What an Amazon search scraper API does
You send a target Amazon search URL. The service loads the page in a real headless browser, rotates the outbound IP, and hands back the rendered response. Your code then either parses the HTML itself or asks the service to extract structured fields for you. The value is not the HTTP call; it is that the browser fleet, proxy pool, and anti-bot handling are someone else's operational problem.
Typical jobs this supports:
- Tracking how a keyword's top results shift over time.
- Comparing price and rating across competitors for the same query.
- Capturing product images and specifications for benchmarking.
- Feeding LLM or analytics pipelines with clean, structured search data.
If your need is a one-off snapshot, a manual export is cheaper. If you need repeatable, geo-specific, scheduled collection, the managed route starts to pay for itself in engineering time.
What goes into a request
At minimum you supply the Amazon search URL and your API key. Beyond that, a small set of parameters covers most real-world needs:
- render_js — run the page in a headless browser and return post-render HTML. Needed when results are injected client-side.
- stealth_proxy — escalate to stealthier proxies when a target is aggressive about blocking.
- country_code — route through a residential IP in a specific country so results reflect that marketplace.
- extract_rules — CSS or XPath selectors that pull specific fields into structured output.
- ai_extract_rules — natural-language extraction instructions when you would rather describe the fields than write selectors.
A dashboard request builder lets you paste a URL, set these parameters, and copy a generated snippet in Python, Node, Go, PHP, Ruby, or cURL, so you can test before committing to a code path.
What comes back
A search-result page carries more than a list of titles. The fields that matter for pricing and competitive work include:
- ASIN and position in the result list.
- Sponsored flag, so organic and paid placements can be separated.
- Price and MSRP.
- Rating and review count.
- Prime eligibility.
- Main image URL.
You can receive these as structured JSON via extract_rules or ai_extract_rules, or as raw post-render HTML if you prefer to parse yourself. Markdown output is also available for LLM-ready pipelines.
A worked example: one keyword, one country
Suppose you want the top results for a keyword on the German marketplace. You would send the Amazon.de search URL with country_code set to DE, render_js enabled, and an extract_rules or ai_extract_rules block describing the fields you want. The response gives you a list of ASINs with position, price, rating, review count, Prime flag, and image URL. You store that snapshot with a timestamp. On the next run, you diff positions and prices to see what moved.
The same pattern scales to multiple keywords and countries by looping over URL/country pairs. The important discipline is to keep the raw response alongside the parsed fields, so a parsing change does not force a re-scrape.
How credits stack
Credits are the unit of consumption, and the cost of a request depends on which features you enable. The base cost is one credit per request. Enabling JavaScript rendering raises the cost to five credits per successful request. Enabling premium proxies costs ten credits without JS rendering, or twenty-five with it. Credits are valid for one month and do not roll over.
That stacking matters when you plan volume. A pipeline that renders JS and uses premium proxies consumes twenty-five credits per successful request, not one. Before you size a plan, multiply your expected request count by the per-request credit cost of your actual configuration, not the base rate.
Three different failure modes, three different fixes
It helps to separate the things that can go wrong, because the fix is different for each:
- Concurrency limits. The API allows a fixed number of simultaneous requests per plan. If you exceed it, requests wait or fail at the API layer. Fix: reduce parallelism or move to a higher-concurrency plan.
- Per-target throttling. Amazon may slow responses for aggressive patterns even when your API concurrency is fine. Fix: space out requests or use multiple sessions.
- Blocking. The target serves a challenge page instead of results. Fix: enable stealth_proxy, which routes through stealthier proxies and a real headless browser.
Treating all three as "rate limiting" leads to the wrong remedy. Check which layer produced the error before changing your request pattern.
When a managed API is not the right tool
A managed API is a poor fit if you need only a handful of pages once, if your compliance team requires you to control the egress IPs directly, or if your data volume is so small that a manual export covers it. It is also a poor fit if you need to interact with a logged-in Amazon account in ways the API does not expose.
For everything else — recurring collection across keywords and countries, with browser and proxy management handled for you — the trade is engineering time for a predictable per-request cost.
A short planning checklist
Before you commit to a plan, work through these:
- List the keywords and countries you need. Each URL/country pair is one request.
- Decide whether you need JS rendering. If the results are server-rendered, you can skip it and save credits.
- Decide whether you need premium proxies. Start without them; escalate only if you see blocking.
- Multiply request count by the per-request credit cost of your configuration.
- Check the concurrency limit against your desired parallelism.
- Confirm whether you need team management, priority support, or a dedicated account manager; those are plan-tier features.
- Run a small pilot before scaling, and keep raw responses for re-parsing.
| Plan | Monthly price | API credits | Concurrent requests | Priority support | Account manager | Team management |
|---|---|---|---|---|---|---|
| Hobby | $19 /mo | 75,000 | 25 | Not included | Not included | Not included |
| Freelance | $49 /mo | 250,000 | 50 | Not included | Not included | Not included |
| Startup (Recommended) | $99 /mo | 1,000,000 | 100 | Not included | Not included | Not included |
| Business | $249 /mo | 3,000,000 | 200 | Included | Included | Included |
| Business + | $599 /mo | 8,000,000 | 400 | Included | Included | Included |
| Enterprise 14 | $999 /mo | 14,000,000 | 500 | Included | Included | Included |
| Enterprise 24 | $1,599 /mo | 24,000,000 | 600 | Included | Included | Included |
| Enterprise 41 (Recommended) | $2,399 /mo | 41,000,000 | 700 | Included | Included | Included |
| Enterprise 69 | $3,599 /mo | 69,000,000 | 800 | Included | Included | Included |
| Enterprise 120 | $5,799 /mo | 120,000,000 | 900 | Included | Included | Included |
Prices are exclusive of VAT. Credits are valid for one month and do not roll over. Requests start at one credit each; JS rendering and premium proxies increase per-request cost. Plan names and prices are a dated reference; check current conditions before purchase.
| Configuration | Credits per successful request |
|---|---|
| Base request | 1 |
| JS rendering enabled | 5 |
| Premium proxies, JS rendering disabled | 10 |
| Premium proxies, JS rendering enabled | 25 |
Costs apply per successful request. Credits are valid for one month and do not roll over.
Frequently asked questions
How do I start scraping Amazon Search?
Create an account, take the API key, and send a GET request with the target Amazon search URL. The service renders the page, rotates proxies, and returns the response. A free trial with 1,000 credits lets you test before choosing a plan.
Will I get blocked when scraping Amazon Search?
Requests go through rotating residential proxies and a real headless browser, so they resemble normal traffic. If a target becomes aggressive, enable stealth_proxy on the request to escalate.
Can I scrape Amazon Search without writing code?
Yes. The dashboard request builder lets you paste a URL, set parameters, and copy a generated snippet in Python, Node, Go, PHP, Ruby, or cURL.
How do I handle JavaScript when scraping Amazon Search?
Pass render_js=true. The page runs in a real headless browser, JavaScript executes server-side, and you receive the post-render HTML.
Can I scrape Amazon Search from a specific country?
Yes. Pass country_code on the request to route through a residential IP in that country. This matters when content varies by geo.
What's the rate limit for scraping Amazon Search?
The API enforces concurrency limits per plan, not per-target rate limits. If the target throttles aggressive patterns, space out requests or use multiple sessions.
How do credits stack when I enable multiple features?
The base cost is one credit per request. JS rendering raises it to five. Premium proxies cost ten without JS rendering or twenty-five with it. Credits are valid for one month and do not roll over.