When is the official API cheaper than scraping?
A target-by-target guide to official API cost, access, returned data, and the cases where scraping is still a different product.
The dataset behind this guide
What is calculated and what is checked. This guide is not a market ranking, so it carries no computed winner: the site's market band prices 100,000 JavaScript-rendered pages a month, and quoting it here would answer a question this page never asked. The prose below is written by a person. Each money figure is checked against the dated vendor price stored with this guide, but that does not mean the vendor page was freshly rechecked. A mismatch must be explained by the shown arithmetic or another cited vendor fact, or the page is not published.
For 10,000 public GitHub resource reads, the official route can cost $0: GitHub Free is $0 per month and an authenticated user receives 5,000 REST requests per hour, so two hour-long quota windows cover the workload. That is before counting the engineering saved by receiving documented JSON instead of page HTML. (GitHub pricing and REST API limits, retrieved 2026-07-26.)
The important test is not simply whether an official API exists. It is whether that API returns the data your job needs, permits your use, and is available to your account. For several targets, the official interface is plainly the best route. For others, the similarly named API manages a seller’s own account and is not a substitute for public-page collection.
A worked comparison: GitHub metadata
Assume the job needs repository, issue, release, or user metadata that GitHub’s REST API exposes. It needs 10,000 reads, can run across two one-hour windows, and does not need the rendered GitHub interface. The official cash calculation is:
2 windows × 5,000 authenticated requests per hour = 10,000 requests
GitHub Free account price = $0 per month
Those are published limits and prices, not a throughput measurement. (GitHub pricing and REST API limits, retrieved 2026-07-26.)
For contrast, the table below assumes one successful basic scrape per public page, without JavaScript, premium proxies, or target multipliers. These are cash floors for a page-fetching product, not equivalent GitHub API responses and not claims that every page will succeed.
A worked comparison: GitHub metadata table
| Scraping route for 10,000 page requests | Published cash floor |
|---|---|
| Bright Data Web Unlocker and Web Scraper API | $15 on pay as you go: 10,000 / 1,000 × $1.50; Bright Data separately lists a 5,000-request free tier. (pricing, retrieved 2026-07-26.) |
| Decodo Web Scraping API | $19 per month for up to 38,000 standard-proxy requests; its free plan stops at 2,000. (pricing, retrieved 2026-07-26.) |
| Scrape.do | $29 per month for 250,000 successful API credits on Hobby. (pricing, retrieved 2026-07-26.) |
| Scrapfly | $30 per month for 200,000 credits on Discovery. (pricing, retrieved 2026-07-26.) |
| ScraperAPI | $49 per month for 100,000 credits on Hobby after its 7-day trial. (pricing, retrieved 2026-07-26.) |
| ScrapingBee | $49 per month for 250,000 credits on Freelance. (pricing, retrieved 2026-07-26.) |
| Oxylabs Web Scraper API | $49 per month on Micro; other-site results are listed at $1.15 per 1,000 within that plan. (pricing, retrieved 2026-07-26.) |
| ZenRows | $69.99 per month for 250,000 basic results on Developer. (pricing, retrieved 2026-07-26.) |
The result is not that scraping products are overpriced. They solve another problem: retrieving web pages through blocks, rendering, and proxy rotation. For this named GitHub job, those capabilities add cost while replacing a structured, supported response with HTML. Use the official API.
What every target’s official API costs and returns
“No separate price published” below means exactly that. It does not mean the surrounding account, commercial agreement, attribution requirement, or engineering work is free.
What every target’s official API costs and returns table
| Target and official API | What it returns | Cost, access, and the honest route |
|---|---|---|
| Amazon Creators API | Product search, item details, variations, and browse-node hierarchy for affiliate shopping experiences. | Amazon publishes no separate per-call price. Access requires Associates enrollment and at least 10 qualifying sales in the trailing 30 days. Eligible credentials initially receive up to 1 request per second and 8,640 requests per day. Use it for compliant affiliate catalog data, not as a general Amazon-page mirror. (introduction and API rates, retrieved 2026-07-26.) |
| Booking.com Demand API | Accommodation, car-rental, attraction, order, payment, and messaging resources made available to a partner. | No public per-call price is published. It requires a managed Affiliate Partner account, an agreed contract, an affiliate ID, and an API token. Use it when you are an accepted Booking.com partner; it is not open public-site access. (prerequisites, retrieved 2026-07-26.) |
| Crunchbase API | Basic access has organization search, organization lookup, and autocomplete with a limited field set. Other endpoints and fields require Advanced or Commercial access. | The documentation does not publish a general per-call price for broader access. Use Basic if its limited organization fields answer the job; otherwise obtain a data license rather than assuming page scraping is equivalent. (Basic APIs, retrieved 2026-07-26.) |
| eBay Browse API | Listing search by keyword, category, GTIN, product, image, charity, or compatibility, plus item details and compatibility checks. | No separate per-call price is published; an application token and production approval or policies apply. For supported listing discovery, this is better than parsing eBay result pages. (Browse API, retrieved 2026-07-26.) |
| Etsy Open API v3 | Inventory, listings, sales orders, and shop-management data, with public and OAuth-authorized endpoints. | No separate request price is published. Seller Apps are scoped to the owner’s shop; broader Personal and Commercial access is reviewed, and Etsy says screen scraping is not allowed. Use the API for an Etsy integration. (Open API documentation, retrieved 2026-07-26.) |
| Meta Graph API for Facebook | Objects and connections exposed to an approved app, such as authorized users, Pages, posts, and related fields. | No separate per-request price is published. Permissions, user or Page authorization, app review, and rate limits define the accessible fields. It is not an unrestricted public-Facebook API. (Graph API overview, retrieved 2026-07-26.) |
| GitHub REST and GraphQL APIs | Structured repository, issue, pull-request, release, user, organization, search, and other platform resources, subject to endpoint permissions. | GitHub Free is $0 per month. REST allows 60 unauthenticated public-data requests per hour and generally 5,000 authenticated requests per hour. Use the API whenever its resources match the job. (pricing and rate limits, retrieved 2026-07-26.) |
| Google Maps Places API | Place search, details, IDs, autocomplete, photos, contact fields, and atmosphere fields, depending on the requested field mask and SKU. | Pricing is field-sensitive. The global list gives Text Search Pro 5,000 free monthly events, then $32 per 1,000 through 100,000; requesting higher-tier fields selects the highest applicable SKU. Use Places when its licensed fields are the job. (usage and field masks and pricing, retrieved 2026-07-26.) |
| Google Custom Search JSON API | JSON web and image results from a configured Programmable Search Engine. | This is not a route for a new buyer. Google closed it to new customers; existing customers get 100 queries per day free, then pay $5 per 1,000, with discontinuation on January 1, 2027. Existing users should plan a migration. (overview and pricing, retrieved 2026-07-26.) |
| Google Merchant API | A merchant’s own Merchant Center products, promotions, inventories, reports, data sources, and account quotas. | No separate API fee is published. It retrieves and manages the caller’s Merchant Center resources, not public Google Shopping search results. Use it for your own feed operations; use another lawful source when the job is market-wide Shopping results. (Merchant API reference, retrieved 2026-07-26.) |
| Idealista Search API | Property information published on Idealista for integration into an approved site or app. | No public per-call price is shown. Idealista asks applicants to describe the project before it issues a key. Request access first; do not budget this as a self-service API. (access request, retrieved 2026-07-26.) |
| Indeed Job Sync API | Create, update, inspect, list, and expire the approved partner’s job postings. It does not return public job-search results. | Indeed says Job Sync calls are free, but the integration is for Indeed partners and an app is provisioned through the Partner Console. Use it to publish and manage jobs, not to replace an Indeed search scraper. (Job Sync guide, retrieved 2026-07-26.) |
| Instagram API with Instagram Login | Authorized professional-account profile, media, comments, and other supported professional-account resources. | No separate request fee is published. It is for professional accounts and authorized app use cases, not general extraction of arbitrary public profiles. Use it when the account owner participates. (Instagram Platform, retrieved 2026-07-26.) |
| LinkedIn APIs | Approved products expose advertising, company, community-management, event, lead, and other data associated with authorized members and administered accounts. | No general per-request price is published. Community Management Development access is limited to 500 app calls and 100 calls per member per 24 hours; many products require review, and Standard access is a separate application. It is not a public people-search API. (access tiers, retrieved 2026-07-26.) |
| Product Hunt API 2.0 | A GraphQL interface with read-only public scope by default and possible approved write scope. |
No public fee is listed, but access requires a token, fair-use limits apply, and default use must not be commercial. A business must contact Product Hunt. Use it for permitted public-scope projects, not an unapproved commercial feed. (API documentation, retrieved 2026-07-26.) |
| Reddit Data API | Authenticated Reddit posts, comments, communities, users, and actions exposed by approved endpoints and scopes. | Reddit publishes no universal commercial rate. Its terms reserve the right to charge and require a separate agreement for commercial use, research above limits, or uses outside the standard terms. Approval and purpose matter more than a nominal $0 entry price. (Data API Terms, retrieved 2026-07-26.) |
| Shopify Storefront API | GraphQL products, collections, carts, and storefront data for one specific Shopify shop. | No separate Storefront API request price is published. A token belongs to a particular store, and Shopify permits at most 100 active storefronts and access tokens per shop. Use it with merchant access; it is not a universal API for arbitrary Shopify stores. (Storefront API guide, retrieved 2026-07-26.) |
| WordPress REST API | JSON posts, pages, taxonomies, users, and other registered WordPress data types; authenticated clients can also create or modify permitted content. | WordPress publishes no separate REST API fee. Public endpoints depend on the site’s configuration, and a generic static site has no equivalent universal API. Try /wp-json/ first on a WordPress target when the needed content is exposed. (REST API Handbook, retrieved 2026-07-26.) |
| TikTok Research API | Public account, content, and Shop data made available through TikTok’s research tools. | No fee is published, but access is restricted to qualifying researchers in named regions and organizations, with noncommercial or public-interest requirements and an application review. It is not a commercial TikTok data API. (Research API, retrieved 2026-07-26.) |
| Tripadvisor Content API | Data covering more than 7.5 million locations, 1 billion reviews and opinions, and 29 languages, plus location search and details. | The first 5,000 calls each month are free after card-backed signup; Search APIs allow up to 10,000 calls per day, and overage pricing is shown at checkout rather than publicly in the FAQ. Tripadvisor also says a newer Terra platform is coming. Under 5,000 comparable calls, use the official route. (Content API FAQ, retrieved 2026-07-26.) |
| Walmart Marketplace APIs | A seller’s or approved solution provider’s items, inventory, orders, pricing, promotions, reports, fulfillment, and related operations. | No separate per-call fee is published. OAuth credentials are tied to Marketplace participation, and tokens last 15 minutes. This manages a seller business; it does not return public Walmart search pages or the full retail catalog as a shopper sees them. (Marketplace API introduction, retrieved 2026-07-26.) |
| X API | Posts, users, direct-message events, relationships, lists, Spaces, communities, likes, and other endpoint resources. | X now publishes pay-per-use prices: $0.005 per Post read, $0.010 per User read, and a 2 million Post-read monthly cap on pay-per-use. Because billing is per resource returned, do not compare one API response with one scraped page as if they were the same unit. (X API pricing, retrieved 2026-07-26.) |
| Yelp Places API | Business search, matching, details, reviews, categories, and other licensed Places fields, depending on access. | The Starter trial limit is 300 calls per 24 hours. Yelp does not publish a general self-service price for higher access on this page; increases depend on a launched integration and review. For a qualifying local-business product inside that allowance, start official. (rate limits, retrieved 2026-07-26.) |
| YouTube Data API v3 | Structured channels, videos, playlists, search results, comments, activities, subscriptions, and authorized write operations. It does not provide a general transcript endpoint. | Google publishes quota rather than a per-call dollar fee. New projects receive 100 search.list calls and 10,000 combined daily units for other endpoints by default; quota extensions require review. Use it for metadata and supported actions, but not when the required output is a transcript. (Data API overview, retrieved 2026-07-26.) |
| Zillow and related approved APIs | Approved components can return valuation, property, mortgage, for-sale posting, professional-review, and directory data. | No general public price is published. Zillow limits access to approved licensees and the reviewed terms specify 1,000 daily calls for Home Valuation and Property Details unless a higher limit is approved. This is licensed data access, not an open public-listing API. (Zillow API terms, retrieved 2026-07-26.) |
Amazon: PA-API comparisons are now obsolete
Amazon’s current first-party position is unambiguous: Product Advertising API 5.0 has been deprecated and is being replaced by the Creators API. The old PA-API documentation says it is no longer maintained, and the current deprecation notice says calls to the deprecated endpoint can receive HTTP 403 AccessDeniedException. A comparison that still tells a new integration to choose PA-API 5 is out of date. (PA-API deprecation notice and old PA-API FAQ, retrieved 2026-07-26.)
Creators API keeps the familiar catalog jobs but changes the integration. It uses OAuth 2.0 bearer tokens instead of AWS Signature Version 4, moves to creatorsapi.amazon endpoints, and uses lower-camel-case request fields. Existing PA-API AWS credentials do not work with it. (migration guide, retrieved 2026-07-26.)
The honest recommendation is narrower than “Amazon has a free API.” Use Creators API if the product is an Amazon affiliate shopping experience, the account has 10 qualifying sales in the last 30 days, and catalog fields meet the need. Use a scraping product only when the required data is genuinely the public page rather than the affiliate catalog response, and only after checking Amazon’s applicable terms. A scraping subscription does not remove the Creators API eligibility rule because it provides a different route and a different data contract. (Creators API introduction, retrieved 2026-07-26.)
Where the official route is unavailable or the wrong product
Three named targets in this site’s target set have no current self-service listing or search API that could be verified. Airbnb directs connectivity through selected software partners, Redfin publishes no current public listing API in the reviewed first-party material, and Glassdoor’s site retains historical API terms without establishing a current self-service API. That is not permission to scrape. It means the next honest step is a partner, license, alternative dataset, or a terms review, not a fabricated official-API price. (Airbnb software partners, Redfin terms, and Glassdoor terms history, retrieved 2026-07-26.)
Two generic target classes also have no universal official API figure. The Generic JavaScript single-page app and Site behind Cloudflare records set official API availability to false and contain no official-API source URL, checked 2026-07-26. These labels describe site setup or delivery classes, not named data owners. The endpoint, access terms, and price can only come from the owner of the particular site, so this guide records the answer as unknown instead of treating the missing price as $0.
The most common false equivalences are now explicit:
- Indeed Job Sync publishes your jobs; it does not search everybody’s jobs. (Indeed Job Sync, retrieved 2026-07-26.)
- Google Merchant API manages your Merchant Center catalog; it does not reproduce public Google Shopping rankings. (Merchant API, retrieved 2026-07-26.)
- Walmart Marketplace APIs manage a participating seller; they do not reproduce the shopper-facing Walmart site. (Walmart Marketplace APIs, retrieved 2026-07-26.)
- Shopify Storefront API is store-specific and merchant-authorized; it is not cross-store discovery. (Shopify Storefront API, retrieved 2026-07-26.)
- YouTube Data API returns supported metadata and actions, not general transcripts. (YouTube Data API, retrieved 2026-07-26.)
- Facebook, Instagram, LinkedIn, TikTok, Reddit, Zillow, Booking.com, and Crunchbase all limit meaningful access by authorization, approval, purpose, license, or commercial agreement. Calling them “free alternatives” hides the decisive constraint. (Meta Graph API, Instagram Platform, LinkedIn access, TikTok Research API, Reddit Data API Terms, Zillow terms, Booking.com prerequisites, and Crunchbase Basic, retrieved 2026-07-26.)
Which route should you take?
Use the official API when your job is GitHub metadata, YouTube metadata, a participating eBay or Etsy integration, an authorized WordPress site, eligible Amazon affiliate catalog access, Places data inside Google’s licensed field model, or Tripadvisor content inside its included allowance. These routes are structured, documented, and in several named cases have a $0 entry cost or allowance. (GitHub, YouTube, eBay, Etsy, WordPress, Amazon, Google Maps, and Tripadvisor, retrieved 2026-07-26.)
Use the account-specific API when the job is to manage your own Indeed jobs, Merchant Center feed, Walmart seller operation, Shopify storefront, Facebook Page, Instagram professional account, or LinkedIn ad or company account. Do not buy a scraper to imitate a supported write or account-management workflow. (Indeed, Google Merchant API, Walmart, Shopify, Facebook, Instagram, and LinkedIn, retrieved 2026-07-26.)
Consider a scraping API when the required output is the rendered public page, ranking, layout, or field that the official interface genuinely does not return. First confirm that the collection and intended use are permitted. Then price the actual target multiplier, rendering mode, proxy class, retry policy, and billing definition.
If the official API is restricted and your use is outside its approved purpose, the answer is not automatically “scrape it.” Sometimes the correct route is a commercial data license, a partner program, a different source, or no collection at all. That restraint is part of the price comparison.
Sources
- docs.github.com/en/rest/using-the-rest-api/rate-limits-for-the-rest-api
- github.com/pricing
- brightdata.com/pricing/web-unlocker
- decodo.com/scraping/web/pricing
- scrape.do/pricing/
- scrapfly.io/pricing
- www.scraperapi.com/pricing/
- www.scrapingbee.com/pricing/
- oxylabs.io/products/scraper-api/web/pricings
- docs.zenrows.com/first-steps/pricing
- affiliate-program.amazon.com/creatorsapi/docs/en-us/introduction
- affiliate-program.amazon.com/creatorsapi/docs/en-us/paapiv5-deprecation
- affiliate-program.amazon.com/creatorsapi/docs/en-us/migrating-to-creatorsapi-from-paapi
- affiliate-program.amazon.com/creatorsapi/docs/en-us/concepts/api-rates
- webservices.amazon.com/paapi5/documentation/faq.html
- developers.booking.com/demand/docs/getting-started/prerequisites
- data.crunchbase.com/docs/crunchbase-basic-using-api
- developer.ebay.com/develop/api/buy/browse_api
- developers.etsy.com/documentation/
- developers.facebook.com/docs/graph-api/overview/
- developers.google.com/maps/documentation/places/web-service/usage-and-billing
- developers.google.com/maps/billing-and-pricing/pricing
- developers.google.com/custom-search/v1/overview
- developers.google.com/merchant/api/reference/rest
- developers.idealista.com/access-request
- docs.indeed.com/job-sync-api/job-sync-api-guide
- developers.facebook.com/docs/instagram-platform/
- learn.microsoft.com/en-us/linkedin/marketing/increasing-access?view=li-lms-2026-05
- api.producthunt.com/v2/docs
- redditinc.com/policies/data-api-terms
- shopify.dev/docs/storefronts/headless/building-with-the-storefront-api/getting-started
- developer.wordpress.org/rest-api/
- developers.tiktok.com/products/research-api/
- tripadvisor-content-api.readme.io/reference/faq
- developer.walmart.com/us-marketplace/docs/introduction-to-marketplace-apis
- docs.x.com/x-api/getting-started/pricing
- docs.developer.yelp.com/docs/places-rate-limiting
- developers.google.com/youtube/v3/getting-started
- www.zillowgroup.com/developers/terms/
- www.airbnb.com/help/article/2908
- www.glassdoor.com/about/terms-history/
- www.redfin.com/about/terms-of-use