scrapeprices 
Scrape.do
The clearest rate card in this tier and one of the few that prints the whole multiplier ladder in one table: 1 credit plain, 5 rendered, 10 residential, 25 residential and rendered, with a monthly allowance rather than a per-call price. Nothing is feature-gated by plan, the free tier is 1,000 credits a month with no card, and credits are only spent on a successful result - Scrape.do counts 2XX, 400, 404 and 410 as success and charges nothing for a timeout or a block. Two things to read before budgeting. First, the plan is a hard cap: at the limit requests stop until you upgrade or buy more credits, and the price of those extra credits is not published anywhere. Second, the base table is a floor, not the bill. A separate per-domain override charges 10 to 200 credits on named sites - google.* 10, linkedin.com 30, chewy.com 50, shopee.* 100, sainsburys.co.uk 200 - so the sites people actually buy a scraping API for are often several times the headline rate, and only the Scrape.do-Request-Cost response header tells you what a given call really cost. Two of those overrides are worth reading twice. amazon.* is the cheapest line in the whole table at 1 credit, but it is silently rerouted to the Amazon Scraper API and returns structured JSON, so raw Amazon HTML is not what you get. And walmart.com is absent from the override table, so a generic call is 1 credit, while the Walmart Scraper API that returns the raw page is 10 - the same host at two prices depending on which door you use.
Plain page, by volume
| Pages per month | Monthly bill | Per 1,000 | Cheapest route | Allowance unused |
|---|---|---|---|---|
| 1k | $29.00 | $29.00 | Hobby | 249,000 units |
| 10k | $29.00 | $2.90 | Hobby | 240,000 units |
| 100k | $29.00 | $0.290 | Hobby | 150,000 units |
| 1M | $99.00 | $0.099 | Pro | 250,000 units |
What the target site does to the price
Scrape.do's base ladder is scoped in its own words to "untargeted domains (domains not in the per-domain table below)", and a separate table overrides the cost for named sites from 10 to 200 credits. Three of the ten basket sites are in that table. The override replaces the base cost rather than adding to it - "the Cost column reflects what you will be charged whether or not you also passed those parameters" - and if you send stricter parameters "the higher cost applies", but Scrape.do never prints what that higher figure is, so a rendered or residential fetch of an overridden site is left unpriced here rather than guessed.
Dollars per 1,000 pages at 100k a month, read off Scrape.do's own pages for each of the 10 sites in the benchmark basket. The ranked price is the median of these, not the cheapest of them. 100k because below it a plan floor is the whole bill and every target prints the same number.
| Target | Plain | Rendered | Residential proxy | Anti-bot | Markdown | What Scrape.do publishes |
|---|---|---|---|---|---|---|
| amazon.com | $0.290 | not published | not published | not published | not published | Listed at 1 credit, the cheapest entry in the whole per-domain table, but only because "amazon.* | Routed via the Amazon Scraper API | Returns structured JSON". You cannot buy raw Amazon HTML at that price, and no rendered, residential or anti-bot figure is printed for it. source |
| ebay.com | $0.290 | $0.990 | $0.990 | $2.49 | $0.290 | No target-specific price; the rate card applies. |
| walmart.com | $0.290 | $0.990 | $0.990 | $2.49 | $0.290 | No target-specific price; the rate card applies. |
| google.com | $0.990 | not published | $0.990 | not published | $0.990 | "google.* | 10 | super=true by default | Residential proxy automatically applied for reliable SERP access." Residential is the same 10 because the residential pool is what the override already buys. Adding render=true is "stricter parameters" so "the higher cost applies", and that figure is published nowhere. source |
| linkedin.com | $2.49 | not published | not published | not published | $2.49 | "linkedin.com | 30 | Custom infrastructure | Specialized handling required for platform restrictions." Thirty times the base rate, and Scrape.do prints no figure for rendering or a residential pool on top of it. source |
| indeed.com | $0.290 | $0.990 | $0.990 | $2.49 | $0.290 | No target-specific price; the rate card applies. |
| zillow.com | $0.290 | $0.990 | $0.990 | $2.49 | $0.290 | No target-specific price; the rate card applies. |
| yelp.com | $0.290 | $0.990 | $0.990 | $2.49 | $0.290 | No target-specific price; the rate card applies. |
| instagram.com | $0.290 | $0.990 | $0.990 | $2.49 | $0.290 | No target-specific price; the rate card applies. |
| tiktok.com | $0.290 | $0.990 | $0.990 | $2.49 | $0.290 | No target-specific price; the rate card applies. |
| Median, which is what ranks | $0.290 | $0.990 | $0.990 | $2.49 | $0.290 | Across the 10 of 10 sites Scrape.do publishes a price for. |
Across the basket, a plain fetch runs from $0.290 to $2.49 per 1,000. Neither end of that is a price you can plan on, because which end you land on is decided by the site you point it at, not by anything you buy.
How Scrape.do compares
At 10,000 pages a month Scrape.do costs $2.90 per 1,000, 5.1x the cheapest published route and 4.1x AnyAPI.
| Provider | 1k a month | 10k a month | 100k a month | 1M a month |
|---|---|---|---|---|
Z  Zyte API | $0.570 | $0.570 | $0.570 | $0.351 |
A  AnyAPI | $0.700 | $0.700 | $0.700 | $0.700 |
O  Olostep | $3.33 | $2.00 | $0.990 | $0.399 |
| S Scrape.do | $29.00 | $2.90 | $0.290 | $0.099 |
S  ScrapeOwl | $29.00 | $2.90 | $0.290 | $0.099 |
List prices, dollars per 1,000 Plain pages at the cheapest published route to each volume. How this is worked out
Every published plan
| Plan | Price | Included credits | Overage each | Note |
|---|---|---|---|---|
| Free | $0 per month | 1,000 | hard cap | 1,000 successful API credits a month, 5 concurrent requests, no card. "Includes EVERY Scrape.do feature" - residential and mobile proxies, geo-targeting, JS rendering, sticky sessions, unlimited bandwidth. |
| Hobby | $29.00 per month | 250,000 | hard cap | Advertised as $0.11 / 1K credits. 10 concurrent requests. $290/year annually. Requests stop at the limit unless you upgrade or buy extra credits, and the extra-credit rate is not published. |
| Pro | $99.00 per month | 1,250,000 | hard cap | Advertised as $0.08 / 1K credits. 50 concurrent requests. $990/year annually. |
| Business | $249 per month | 3,500,000 | hard cap | Advertised as $0.07 / 1K credits, marked Most Popular. 100 concurrent requests, dedicated support with an account manager. $2,490/year annually. |
| Advanced | $699 per month | 10,000,000 | hard cap | Advertised as $0.06 / 1K credits. Dedicated Slack channel. $6,990/year annually. |
| Enterprise | on request | - | not published | Contact sales. Custom SLA and dedicated infrastructure; no public price, and the annual discount does not apply to it. |
The published base table: "Normal request (datacenter proxy) 1", "Datacenter + headless browser (render=true) 5", "Residential / mobile (super=true) 10", "Residential / mobile + headless browser 25". Anti-bot is recorded as 25 because Scrape.do sells no separate stealth or unblocker product: its documentation says "Some websites require advanced bypass methods (residential proxies, headless browser rendering, or custom anti-detection solutions), which consume more credits per request. For example, a basic datacenter request costs 1 credit, while a residential proxy request with JS rendering costs 25", so super+render at 25 is the hardest configuration on the general rate card. That is a floor, not a ceiling: a separate per-domain table overrides the base cost for named sites, and those run from 10 to 200 credits (google.* 10 with super=true by default, linkedin.com 30, chewy.com 50, shopee.* 100, sainsburys.co.uk 200). Markdown is 1 because output=markdown is a response format, not a cost dimension - it does not appear in the request-costs table, which prices only proxy type and rendering. The authoritative figure for any single call comes back in the Scrape.do-Request-Cost response header.
Cost by kind of fetch
Per 1,000, at 10,000 pages a month.
What it can do
Only what the provider's own pages state.
Where these numbers came from
- https://scrape.do/pricing/ read 2026-09-14 / Plan cards Free $0/1,000, Hobby $29/250,000, Pro $99/1,250,000, Business $249/3,500,000, Advanced $699/10,000,000, Enterprise contact sales, all labelled "Successful API Credits"; "Full Power. Every Plan. No feature gates." listing datacenter proxies, residential and mobile proxies, JavaScript rendering, geo-targeting (160+ countries), sticky sessions, unlimited bandwidth on every tier. FAQ verbatim: "You're only charged when a request returns a successful response with valid content. You won't lose credits on timeouts, blocks, or errors." FAQ verbatim on annual: "Annual billing gives you up to 20% off - pay for 10 months and get 12 - on every plan except Enterprise. Hobby is $290/year instead of $29/month, Pro is $990/year instead of $99/month, Business is $2,490/year instead of $249/month, and Advanced is $6,990/year instead of $699/month." FAQ verbatim on the cap: "once you hit your monthly limit, requests will stop unless you upgrade your plan or purchase additional credits." Footer: "Copyright (c) 2020 - 2026 Scrape.do | PACKEND, LLC".
- https://scrape.do/documentation/request-costs/ read 2026-09-14 / Base cost table 1 / 5 / 10 / 25 for datacenter, datacenter+render, residential-mobile (super=true), residential-mobile+render. "Scrape.do charges credits only for successful results", with 2XX, 400, 404 and 410 counted as successful. Per-domain override table from 10 to 200 credits: amazon.* 1 via the Amazon Scraper API, google.* 10 with super=true by default, linkedin.com 30, chewy.com 50, shopee.* 100, sainsburys.co.uk 200. Scrape.do-Request-Cost response header is the authoritative per-call cost. No cost row for output=markdown, geoCode or regionalGeoCode.
- https://scrape.do/documentation/ read 2026-09-14 / Proxy Mode (use it as an outbound proxy, not only as an API); geoCode "Choose the right country for your target web page" and regionalGeoCode for continent-level targeting; output parameter for "raw or markdown format"; Asynchronous API with webhooks (the old callback parameter is marked Deprecated); dedicated Amazon, Google, Walmart, YouTube and ChatGPT scraper endpoints. No uptime SLA is published outside the Enterprise tier.
- https://scrape.do/privacy-policy/ read 2026-09-14 / "Welcome to PACKEND, LLC. PACKEND, LLC operates Scrape.do", and the data-transfer clause names the United States as the place of processing. Used for the operating company and country only.
Something out of date? Send the provider's own page to hello@scrapeprices.com and it is corrected on the next capture.