Niches · Web scraping & crawling
Structured records from a named platform without writing selectors
Four or more companies' docs show a full fit.
What counts as done
A JSON record with the platform's named fields (price, rating, address, title) in one call or one Actor run, with a per-result price known before the call.
How agents ask for it: get this Amazon product's price and reviews
, pull the Google results for this query as JSON
, list businesses on Google Maps for this area
, give me this LinkedIn profile as fields
Does the job 14
The docs show it meets the test above, and nothing we read says otherwise.
| Listing | Why |
|---|---|
| Bright Data C Bright Data Ltd. | search_scrapers, get_scraper_details (input fields, sample_input, output fields) and run_scraper cover thousands of site scrapers billed per record; a run over 45 seconds returns a snapshot_id to poll with get_scraper_progress. acc-26 replaced an Amazon scraper with it. source |
| Crawl4AI D CONTEXT4AI PTE LTD | Recipes such as amazon-product, amazon-reviews, ebay-product and linkedin-profile return rows with fixed named columns; GET /recipes lists inputs, columns and the cost line before you run one, and an empty result bills no records. source |
| Decodo Web Data API B UAB "Data troops" | Target templates (amazon_product, amazon_search, google_search and others) with parse: true return typed JSON, per-1,000 prices are published per plan, and parsed-result codes 12002 to 12009 flag partial or failed parsing. source |
| Diffbot C Diffbot Technologies Corp. | Product API returns offerPrice, regularPrice, sku, brand, availability, specs and reviews from any e-commerce page for 1 credit, and Analyze classifies a URL and extracts the matching schema in one call. source |
| Oxylabs Web Scraper API B oxylabs, UAB | POST /v1/scrape/{target}/{collection} such as /amazon/product or /google/search takes an ASIN or query and returns typed fields under results[].json; pricing lists Amazon from $0.50 and Google from $1.00 per 1,000 results on Starter. source |
| Qovia D Qovia | Each endpoint returns typed JSON for one platform at a published per-call price, meta.billed and charged_usd report the cost, empty results aren't charged, and lists page with a cursor. Covers Douyin, Xiaohongshu, Weibo, TikTok, YouTube, X; Amazon and Shopee are coming soon. source |
| Scrape.do E PACKEND, LLC | Amazon Scraper API plugins (PDP, has, search) return typed JSON at 1 credit; Google Search, Maps, Shopping and others at 10 credits a request, stated per endpoint. LinkedIn is raw HTML at 30 credits. source |
| ScrapeAtlas B OD. Media & Technology LLC | Typed JSON per platform: Amazon product with price and rating, LinkedIn profile fields, 50 platforms. One credit per successful request, status success, partial or failed in every body. No Google or Google Maps. source |
| ScraperAPI D ScraperAPI LLC | Structured Data Endpoints for Amazon, eBay, Walmart, Home Depot, Google SERP, Google Maps and Redfin return JSON or CSV, sync or async, with costs on the credits page and max_cost on each call. source |
| Scrapewise B BEBOTECH OÜ | get_amazon_product (1 credit), search_amazon (1 a page), google_search (10), find_local_businesses (10 a call) and idealo_offers return typed fields; every paid tool takes quote_only and reports its cost. source |
| ScrapingBee B VostokInc SAS | Google Search, Amazon Search, Product and Pricing, Walmart and YouTube endpoints return typed fields (asin, price, rating) at stated credits, 5 light or 15 rendered for Amazon Product. source |
| Scrapingdog E OpenData Labs Private Limited | Google Search and Maps (5), Amazon product, search and has, eBay, Walmart and LinkedIn profiles return JSON at credits listed per API; amazon_reviews is marked unavailable. source |
| SoldFetch E SoldFetch | Sold and active eBay listing search, category search and item details return typed JSON for ebay.com, .co.uk and .de at one unit a successful page or lookup; failed requests and idempotent replays cost zero. source |
| Zyte API BB Zyte Group Limited | Automatic extraction returns product, productList, article, jobPosting and serp with fixed schemas (price, rating) on any site, priced per tier plus the extraction fee. acc-20 built a price alert on it without selectors. source |
Does the job, with a limit 15
It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.
| Listing | Why |
|---|---|
| Apify MCP Server A Apify Technologies s.r.o. | fetch-actor-details shows the input schema and price before call-actor, Google Maps Scraper lists from $1.50 per 1,000 places, and maxTotalChargeUsd or maxItems cap a run. acc-25 saw Amazon Product Scraper succeed 66 per cent and failed runs still cost credits. source |
| BrowserAct D Hongkong Ziniao Technology Co., Limited | Official templates listed by GET /v3/bots/templates and agent-built bots return typed output files, GET /v3/bots/{id} gives the input_schema and the task record shows credits used, but the price before a run is an estimate range and template names weren't readable. source |
| Context.dev A Context Dev Inc. | No Amazon, Google or Maps templates. The product format returns typed product fields from any page for 1 extra credit and json extraction takes a schema for 4; People and Company Search are priced per page. source |
| fastCRW B Nanofy OÜ | No first-party Amazon or Google parsers. crw_find_data lists third-party data providers with creditsCost before the call and crw_get_data runs one (proxy mode only); LLM extract costs 5 credits a page. source |
| Firecrawl MCP A SideGuide Technologies, Inc. (d/b/a Firecrawl) | No first-party Amazon, Google or Maps parsers. json format extracts a schema with an LLM for 4 extra credits a page, and Alexandria routes some URLs to third-party providers billed at provider prices shown in metadata.provider after the call. source |
| Olostep B Olostep Technologies Inc. | Parsers @olostep/google-search and @olostep/amazon-it-product return typed JSON for 1 to 5 credits, but Amazon covers the Italian storefront only and Google Maps and News parser ids are given out on request. source |
| Outcrawl B Aeonmind LLC (trading as Outcrawl) | No platform parsers. An agent run with a JSON schema returns typed records, but it is billed by model usage, browser time and proxy at about 15 credits a minute under caps.budget, not per result. source |
| ScrapeGraphAI B ScrapeGraphAI, Inc. | extract runs an LLM with a prompt or JSON schema at 5 credits; no ready parsers for Amazon, Google or LinkedIn, so fields depend on the prompt. source |
| Scrapeless C NST LABS TECH LTD. | Amazon product returns buybox_prices and rating and Google Search is parsed, but runs return a taskId to poll and the per-request price isn't on the API pages (the pricing page renders by script). source |
| ScrapeOps Proxy API Aggregator D ScrapeOps Limited | auto_extract=amazon and other maintained parsers return JSON from the aggregator, but their credit cost isn't in the cost table; the typed Data APIs (Amazon, eBay, Walmart) are a separate product. source |
| Scrapfly BB Joam Intelligence, LLC | No Amazon, Google or LinkedIn parsers; the Extraction API takes a JSON schema or LLM prompt per page, so fields depend on the prompt. source |
| ScrapingAnt E DATAANT SP. Z O.O. | No platform parsers. The AI extractor takes a prompt and bills 1 credit per 30 characters of page Markdown plus output, so the price is unknown before the call. source |
| Spider BB BAGELMEN LLC | No Amazon, Google or LinkedIn parsers; css_extraction_map needs selectors and the prompt-driven /ai routes need an AI plan from $6 a month. source |
| WebScraping.AI B Urlooker LLC | /data has dedicated parsers for Amazon, LinkedIn, Instagram, YouTube and others and learns other sites, but its price varies by site and parse_failed pages are billed; /serp is a fixed 15 credits. source |
| ZenRows A ZENROWS, S.L. | extract=auto returns typed JSON for prepared domains listed by GET /v1/extract/domains (amazon.com among them) at the Fetch credit weight the request needed, 1 to 25; unprepared domains return REQS007. source |
Doesn't do this job 3
Listings in the category that do something else, with the reason.
| Listing | Why |
|---|---|
| jevcrawl B Rayen Mlayeh | No platform parsers or typed fields; results are verbatim passages with source URLs. source |
| Jina Reader C Jina AI GmbH | No platform parsers or typed endpoints; output is Markdown or JSON of the page text. ReaderLM extraction is a separate model, not a per-platform schema. source |
| Microlink B Microlink Labs, S.L. | No platform parsers; data rules take CSS selectors you write, and meta returns normalised page metadata only. source |
1 listings in the category were added after this reading and aren't marked yet.
What stops agents finishing
- Actor reports SUCCEEDED with an empty dataset and still bills the start fee
- platform template broken by a dependency and failing one run in three
- run exceeds the MCP tool wait so the agent must poll separately
Around it
Before: search.find-url. After: spreadsheets.append-row, crm.enrich.
Where this comes from
Weight 10 of 100 in Web scraping & crawling: 4 of 35 accounts; Apify production stats, Zyte automatic extraction, Bright Data SERP and MCP posts
The same as data: /api/v1/use-cases/web-scrapers.platform-structured-data.json, or the MCP tool get_use_case with id web-scrapers.platform-structured-data. Other jobs in this category: /api/v1/niches?category=web-scrapers.
For companies
Do agents find, use and choose your tools?
An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.