How to Scrape Amazon Prices With Python
Learn which Amazon pricing APIs are documented, how to inspect offer data with Python, and what to check before automating price collection.

Short answer: If you are an eligible Amazon seller or developer, start with Amazon’s documented Selling Partner API (SP-API) Product Pricing API. It is designed for seller pricing workflows. For a specific seller listing, Amazon also documents a legacy v0 getListingOffers operation. These are authenticated APIs, not browser scraping. The sources reviewed for this guide do not establish that automated scraping of Amazon retail pages is permitted, stable, or suitable, so this article does not present a headless-browser recipe as an endorsed method.
Python can help you inspect and process the resulting offer data, but acquiring it requires Amazon authorization, the right role, and request signing according to the current SP-API documentation. The reviewed references do not provide enough detail to give you a complete, reliable authentication implementation here. This guide shows how to choose the documented route, prepare your integration, and safely process a saved API response without inventing credentials or a signing procedure.
1. Choose the data source before writing code
“Amazon price” can mean several different things: a seller’s listing price, shipping, a landed total, the lowest offer, the Featured Offer, or an expected Featured Offer price. These values are not interchangeable. Marketplace, item condition, customer type, and the product or seller listing all affect what the API can return.

| Route | What the reviewed sources establish | What to check |
|---|---|---|
| Product Pricing API, v2022-05-01 | Amazon describes this as a way to retrieve catalog pricing and offer information for automated seller pricing solutions. It is marked “Sellers only.” | Confirm the operation, required Pricing role, region, inputs, and response schema in the current reference. |
getListingOffers, legacy v0 |
Returns the lowest priced offers for one SKU listing. It requires a seller SKU, marketplace ID, and item condition. | Check whether this legacy operation and its response model fit your current integration before implementing it. |
| Retail-page browser scraping | The reviewed sources do not establish permission, stable page structure, or suitability for automated retail-page collection. | Check Amazon’s current terms and policies independently before considering this route. |
| Affiliate-facing API | A search result mentions an Offers resource with price, availability, and savings, but current access and terms were not confirmed in the reviewed research. | Verify current official affiliate API documentation and eligibility before relying on it. |
The current Product Pricing API overview identifies version v2022-05-01. The getListingOffers reference is under legacy v0, so do not assume its version, inputs, or schema are the same as a current operation. Amazon’s overview also describes pricing notifications as complementary event-driven signals to on-demand pulls. If your use case needs ongoing updates, review those notifications in the official documentation instead of polling blindly.
2. Check eligibility and define the price you need
- Confirm you can use SP-API. Amazon’s onboarding describes SP-API as a REST API for seller or vendor workflows. Its general developer page says creating a developer profile requires a Professional selling account and primary-account-user status. Amazon evaluates profile details.
- Register and authorize the application. Follow Amazon’s current onboarding for developer profile creation, application registration, authorization, and production access. Public applications are authorized by a selling partner; private apps can be self-authorized. Roles control access to operations.
- Confirm the role and region. The current pricing overview lists
getCompetitiveSummaryandgetFeaturedOfferExpectedPriceBatchas requiring the Pricing role in North America, Europe, and the Far East. Verify the exact operation and marketplace region for your account. - Choose identifiers and semantics. Decide whether you need an ASIN-level catalog view or offers for your own SKU listing. For the legacy listing-offers operation, select the marketplace, item condition, and customer type.
- Read the current schema. Check the operation’s required fields and response model immediately before coding. In particular, decide whether you need price only, shipping, landed price, availability, or Featured Offer information.
For the legacy getListingOffers operation, accepted conditions are New, Used, Collectible, Refurbished, and Club. The optional customer type is Consumer or Business, and defaults to Consumer. A result for one condition or marketplace should not be presented as a universal price for every buyer.
3. Python: inspect and process an offer response
SP-API requests require Amazon’s current authorization and signing process. Because the reviewed references do not specify enough detail to construct that process accurately, the example below begins with a response JSON file obtained through an authorized integration. It is runnable as-is once you save a representative response body as listing_offers.json. The script prints the response structure so you can identify the price fields in the schema for your selected operation.
import json
from pathlib import Path
response_path = Path("listing_offers.json")
with response_path.open(encoding="utf-8") as file:
payload = json.load(file)
if not isinstance(payload, dict):
raise ValueError("Expected a JSON object at the response root")
print(json.dumps(payload, indent=2, ensure_ascii=False))
Save the response body, not a screenshot or a log containing access tokens. Response models differ by operation and API version; inspect the documented schema instead of assuming that a field named price always means the same thing. For repeatable collection, save the marketplace, seller SKU or ASIN, condition, customer type, operation name, API version, and retrieval timestamp alongside each parsed record.
Here is a second runnable Python example for a response you have already mapped to a normalized format. It calculates a displayed offer total only when your own schema mapping has explicitly provided both an item amount and shipping amount. It does not guess Amazon’s response field paths.
from decimal import Decimal, InvalidOperation
# Replace these example records with values mapped from the operation's
# documented response schema. Amounts are decimal strings in one currency.
offers = [
{"seller_id": "SELLER_A", "item_amount": "19.99", "shipping_amount": "0.00", "currency": "USD"},
{"seller_id": "SELLER_B", "item_amount": "18.50", "shipping_amount": "3.49", "currency": "USD"},
]
normalized = []
for offer in offers:
try:
item = Decimal(offer["item_amount"])
shipping = Decimal(offer["shipping_amount"])
except (KeyError, InvalidOperation) as exc:
raise ValueError(f"Offer is missing a valid amount: {offer}") from exc
if item < 0 or shipping < 0:
raise ValueError(f"Offer contains a negative amount: {offer}")
normalized.append({**offer, "landed_amount": item + shipping})
currencies = {offer["currency"] for offer in normalized}
if len(currencies) > 1:
raise ValueError("Compare offers in one currency at a time")
for offer in sorted(normalized, key=lambda row: row["landed_amount"]):
print(offer["seller_id"], offer["currency"], offer["landed_amount"])
This example demonstrates data handling, not the semantics of any particular Amazon field. Use the API’s documented response definition to map amounts and currency correctly. If shipping is absent, unknown, or represented differently, preserve that uncertainty rather than treating it as zero.
4. Parameters and response details to get right
Listing-offer inputs
The legacy getListingOffers reference requires SellerSKU, MarketplaceId, and ItemCondition. Its condition values are listed above. CustomerType accepts Consumer or Business, with Consumer as the default. The operation is for a particular seller SKU listing and returns the lowest priced offers; it is not a general promise to return every retail price on Amazon.

Current Product Pricing operations
The current overview names operations including getCompetitiveSummary and getFeaturedOfferExpectedPriceBatch. It describes operation-specific batch use cases: up to 40 SKUs for Featured Offer Expected Price and up to 20 ASINs for featured offers. These are batch sizes for the named use cases, not a universal maximum for every pricing operation. Confirm current inputs, role requirements, regional availability, and response semantics in Amazon’s current reference before choosing one.
Model the values honestly
- Store item price, shipping, and any landed total as distinct fields.
- Keep marketplace, condition, customer type, currency, identifier, and timestamp with each record.
- Represent missing values as missing; do not silently turn them into zero.
- Keep Featured Offer data distinct from the lowest priced offer and from a seller’s own listing price.
- When comparing offers, group by marketplace, condition, customer type, and currency.
5. Rate limits, reliability, and cost
Amazon documents a default usage plan of one request per second with a burst of two for the legacy getListingOffers operation. Amazon notes that the applied limit may be higher for some sellers and that a response rate-limit header may show the effective plan. Treat this as an operation-specific documented default, not a universal SP-API limit. Read the current endpoint reference and response headers for the effective limit that applies to your integration.
Plan for both access failures and temporary service failures. The listing-offers reference lists 401 for malformed or invalid authorization; 403 for access denied, unauthorized, expired token, or invalid signature; 429 for excessive request frequency; and 500 or 503 for service errors. Build handling around the actual response, preserve useful diagnostic identifiers where permitted, and avoid logging secrets. Follow current Amazon guidance for throttling and retry behavior; no specific retry policy is prescribed in the sources reviewed for this article.
Pricing data collection has practical costs beyond API calls: developer setup, authorization, maintenance as schemas and access requirements evolve, storage, and the cadence at which your application needs fresh data. Notifications may complement on-demand pulls for a changing-price workflow. The available research does not establish a general call price, throughput promise, or complete list of costs, so check Amazon’s current terms and account details rather than budgeting from an assumed rate.
6. Troubleshooting
| Symptom | Likely cause | What to check |
|---|---|---|
401 |
Malformed or invalid authorization. | Review the current authentication setup and confirm the request is built for the correct application and account. |
403 |
Access denied, missing authorization, expired token, or invalid signature. | Verify selling-partner authorization, operation role, token state, signing inputs, and region. |
429 |
Requests exceeded the operation’s effective rate limit. | Reduce request frequency, inspect the response rate-limit header, and follow current Amazon throttling guidance. |
500 or 503 |
Amazon service error or temporary unavailability. | Capture the response details safely and apply the retry guidance in the current API documentation. |
| No expected offer or price field | Wrong operation, API version, marketplace, condition, or a mistaken assumption about the response schema. | Confirm the selected operation’s documented model and inspect the raw response before changing your parser. |
| Two “lowest” prices disagree | Different conditions, customer types, marketplaces, shipping treatment, or timestamps. | Compare records with matching context and identify whether each value is item price or landed price. |
7. Browser capture for visual review
If your task is to review how a product page appears, a screenshot can be useful as a visual artifact. A screenshot does not turn page content into a structured, authoritative price feed. Check Amazon’s current policies before automating access to retail pages, and use a documented data source for price records when one fits your eligibility and use case.
For a visual capture workflow, ScreenshotNeo is a website screenshot API and MCP server. Its API takes a URL and returns an image or PDF. The documented options include viewport or full-page capture, device presets, waiting conditions, cookies and headers, and caching. These are capture controls; they do not replace Amazon pricing data or SP-API authorization.
8. Or skip the browser setup
To capture a page visually with one request, use ScreenshotNeo. See the API documentation for request options. Use a page you are authorized to capture.
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webp
import requests
r = requests.get("https://api.screenshotneo.com/v1/shot", params={"access_key": "YOUR_API_KEY", "url": "https://stripe.com"}, timeout=90)
open("shot.webp", "wb").write(r.content)
const q = new URLSearchParams({ access_key: 'YOUR_API_KEY', url: 'https://stripe.com' });
const res = await fetch(`https://api.screenshotneo.com/v1/shot?${q}`);
- Cookie and consent banners, newsletter popups, and chat widgets are removed before the shot; each cleanup step can be turned off.
- Bot checks, blank pages, timeouts, failed loads, and cache hits cost nothing; response headers say which outcome occurred and whether it was billed.
- An MCP server gives AI agents tools to take screenshots, get page information, and capture PDFs.
- The Free plan includes 1,000 screenshots a month with no card. Paid plans start at $5 for 3,000 shots.
See ScreenshotNeo for the service and sign up for 1,000 free screenshots a month, with no card.
9. FAQ
Can I get an Amazon price with one Python request?
You can make an API request after completing Amazon’s required setup, authorization, and signing. The sources reviewed here do not provide enough detail for a trustworthy complete authentication snippet, so use Amazon’s current onboarding and operation reference for that part.
Does the lowest priced offer mean the Featured Offer?
No. The documentation describes distinct pricing concepts and operations. Keep lowest offer, listing price, landed price, and Featured Offer data separate.
Can I use these endpoints for any marketplace or buyer?
Do not assume so. Select the marketplace and relevant customer type, condition, region, and operation based on Amazon’s current documentation and your authorization.
Is a screenshot sufficient for price tracking?
A screenshot records a visual page, not a normalized offer response. Use it for visual review; rely on an authorized data source and explicit field definitions for structured price analysis.


