300 Walmart products, daily
9,000 product pages a month at 5 credits is 45,000 credits. Hobby at $49 covers it with room to spare. Cheaper than any scheduled tool on the meter alone, and still just HTML or JSON landing in your code.
ScraperAPI Pricing in 2026, Every ScraperAPI Plan, Credit Multiplier and Real Monthly Bill
8 min read
ScraperAPI pricing starts at $49 a month for 100,000 API credits and runs to $1,975 a month for 21,500,000, with 10 percent off on annual billing. The plan price is only half the answer: a plain page costs 1 credit, a Walmart or Amazon page 5, a JavaScript-rendered page 10, a Google result 25 and a hard, protected page up to 75. The same $149 plan buys anywhere from 13,000 to 1,000,000 pages.
figures taken from scraperapi.com/pricing and the ScraperAPI documentation on credits and billing
Robot console
Standing by Running · s Complete · rows
1 · Pick a target
3 · Fields to extract
Agent log
Crawl graph
Extracted data · rows
Want this data fresh every morning, without lifting a finger?
The plans
ScraperAPI pricing for every plan
USD per month. Every plan includes JavaScript rendering, premium residential and mobile IPs, JSON auto
parsing, CAPTCHA and anti-bot handling, automatic retries and unlimited bandwidth. What changes between
tiers is the credit allowance, concurrency, geotargeting and overage.
| Plan | Monthly | Annual, per month | API credits | Threads | What else changes |
|---|---|---|---|---|---|
| Free | $0 | n/a | 1,000 | 5 | Plus a 7-day trial with 5,000 credits, no card |
| Hobby | $49 | $44.10 | 100,000 | 20 | Geotargeting US and EU only, 30 days of analytics |
| Startup | $149 | $134.10 | 1,000,000 | 50 | Geotargeting US and EU only, 30 days of analytics |
| Business | $299 | $269.10 | 3,000,000 | 100 | Country-level geotargeting, unlimited analytics history |
| Scaling | $475 | $427.50 | 5,000,000 | 200 | Adds pay as you go past the allowance |
| Professional | $975 | $877.50 | 10,500,000 | 300 | Priority support, pay as you go |
| Advanced | $1,975 | $1,777.50 | 21,500,000 | 500 | Priority routing, pay as you go |
| Enterprise | Custom | Custom | 22,000,000 and up | 500 and up | Dedicated support team, Slack support |
Professional and Advanced were also showing limited bonus credits (250,000 and 500,000) when we checked; treat those as a promotion, not part of the plan. Per 1,000 plain requests, the price falls from $0.49 on Hobby to $0.15 on Startup and about $0.09 on the top tiers, so the jump from Hobby to Startup is the biggest discount on the ladder.
The meter
How ScraperAPI credits are counted per request
The site you hit and the options you switch on set the multiplier. You choose the options; the domain
list is ScraperAPI's.
| Request | Credits | When you need it |
|---|---|---|
| Normal page | 1 | Static HTML sites that do not fight bots |
| E-commerce domain (Amazon, Walmart, eBay) | 5 | Any product or search page on those stores |
| SERP domain (Google, Bing) | 25 | Rank tracking, SERP research |
| Social domain (LinkedIn) | 30 | Public LinkedIn pages |
| render=true or premium=true or screenshot=true | +10 | Pages that build their content with JavaScript, or block datacenter IPs |
| premium=true with render=true | 25 | JavaScript sites that also block ordinary proxies |
| ultra_premium=true | 30 | The hardest sites; paid plans only |
| ultra_premium=true with render=true | 75 | The hardest JavaScript sites; paid plans only |
| Bot protection bypass (Cloudflare, Turnstile, DataDome, PerimeterX) | +10 | Charged when ScraperAPI has to get past it |
Country code, device type, sessions, wait-for-selector and autoparse carry no extra charge. Billing counts responses with status 200 and 404, plus requests you cancel before ScraperAPI's 70-second window ends. That last rule matters more than it looks: set your client timeout under 70 seconds and slow pages get billed even though your code gave up on them.
Pages per plan
How many pages each ScraperAPI plan covers
Credits divided by the multiplier. This is the table to size a plan from, because the credit number on
the pricing page only equals pages for plain, unprotected sites.
| Plan | Plain page (1) | Walmart or Amazon (5) | Rendered (10) | Google result (25) | Ultra and render (75) |
|---|---|---|---|---|---|
| Hobby, $49 | 100,000 | 20,000 | 10,000 | 4,000 | 1,333 |
| Startup, $149 | 1,000,000 | 200,000 | 100,000 | 40,000 | 13,333 |
| Business, $299 | 3,000,000 | 600,000 | 300,000 | 120,000 | 40,000 |
| Scaling, $475 | 5,000,000 | 1,000,000 | 500,000 | 200,000 | 66,666 |
| Professional, $975 | 10,500,000 | 2,100,000 | 1,050,000 | 420,000 | 140,000 |
A page is a request, not a row. One Walmart search page can hold 40 products, so if the listing page carries every field you need, the per-row cost drops a lot. If you need the product page for seller, stock or specifications, you pay per product. The Walmart scraper and Amazon scraper pages list which fields live on which page.
Worked bills
What ScraperAPI costs for four common jobs
Thirty-day months, successful requests only, no retries counted against you.
2,000 JavaScript pages, daily
60,000 rendered pages at 10 credits is 600,000 credits: Startup at $149. Add premium proxies because the sites block datacenter IPs and it becomes 25 credits, 1.5 million credits, and Business at $299.
500 keywords on Google, daily
15,000 result pages at 25 credits is 375,000 credits: Startup at $149. Track ten result pages deep instead of one and it is ten times that, which is Scaling at $475 with pay as you go behind it.
1,000 pages on a hard, protected site, daily
30,000 requests at ultra premium with rendering is 75 credits each, 2.25 million credits: Business at $299. The same page count on an easy site would fit on Hobby, so ask which bucket your targets fall into before you pick a tier.
None of these bills include the developer. ScraperAPI returns the page; turning it into named columns, running it on a schedule, storing history and alerting someone when a number changes is your code. That is the part that decides whether ScraperAPI is cheap for your team.
Compare
ScraperAPI pricing compared with other scraping APIs and WebRobot
Each vendor counts something different, which is why headline prices do not line up.
| Tool | Entry paid plan | What the meter counts | Who does the parsing and scheduling |
|---|---|---|---|
| ScraperAPI | $49, 100,000 credits | Credits per request, 1 to 75 by site and options | You |
| ScrapingBee | $49 Freelance, 250,000 credits | Credits per request, 1 to 75, JavaScript on by default | You (ScrapingBee pricing) |
| Zyte API | Pay as you go | Per 1,000 successful responses, tier set by Zyte per site | You (Zyte pricing) |
| Firecrawl | $19 Hobby, 5,000 credits | Credits per page, +4 for structured JSON | Mostly you (Firecrawl pricing) |
| Bright Data | Pay as you go | $1.50 per 1,000 records or per GB for proxies | Split (Bright Data pricing) |
| WebRobot | $79 Launch, 10,000 records | Rows delivered, whatever the site | We do: named columns, schedule, Sheets or API, alerts on Scale |
Being straight about it: on raw volume ScraperAPI is cheaper. A million plain pages for $149 is a price no scheduled tool matches. WebRobot costs more per unit because a record is a finished row in your sheet, with the scheduler, parser, retries and delivery already done, and the price does not change when a site needs rendering or a login. If you have a developer and a pipeline, ScraperAPI is a good buy. If you have a sheet and a deadline, the web scraping API page shows how our records arrive as JSON, and pricing has the plan limits.
Fit
Which ScraperAPI plan to buy, or when to skip it
Buy Hobby or Startup
A developer, a known list of mostly plain or e-commerce pages, US or EU targets, and a script that already runs. Startup is the value tier; Hobby is for testing volume.
Buy Business or Scaling
You need country-level geotargeting, render most pages, or cannot afford requests stopping at the cap. Scaling is the first tier with pay as you go and a spending cap.
Skip the API
Nobody on the team writes code, the data has to land in a sheet every morning, or someone needs a Slack alert when a competitor price moves. That is what competitor price monitoring on WebRobot is for.
One more line item people forget: the page is not always the data you need. If half your sources are PDFs, scanned invoices or supplier price lists sent as attachments, a web API cannot help with those, and document data extraction is the tool for that half of the job.
FAQ
ScraperAPI pricing questions, answered
ScraperAPI costs $49 a month for 100,000 API credits on Hobby, $149 for 1,000,000 on Startup, $299 for 3,000,000 on Business, $475 for 5,000,000 on Scaling, $975 for 10,500,000 on Professional and $1,975 for 21,500,000 on Advanced. Annual billing takes 10 percent off every tier. Enterprise is quoted by sales from 22,000,000 credits.
There is a free plan with 1,000 API credits a month and up to 5 concurrent connections, which is enough to test a few hundred plain pages. New accounts also get a 7-day trial with 5,000 credits and no card. Anything recurring, JavaScript rendering on a volume of pages, or a protected site needs a paid plan.
Every successful request costs credits, and the number depends on the site and the options. A normal page is 1 credit. Amazon, Walmart and eBay pages are 5. Google and Bing are 25 and LinkedIn is 30. Rendering JavaScript or premium proxies add 10, both together cost 25, and ultra premium with rendering costs 75.
No. The credit balance resets when the subscription renews, so unused credits from a quiet month are gone. You can move up or down a plan from the billing page at any time, which is the practical way to match the allowance to a month that is busier or quieter than usual.
On Hobby, Startup and Business you upgrade to the next tier or ask sales for a custom plan; requests stop until you do. Scaling, Professional, Advanced and Enterprise include pay as you go, which keeps requests running at a fixed per-credit rate, with a monthly spending cap you set in the dashboard.
Mostly no. ScraperAPI bills requests that return 200 or 404, plus requests you cancel before its 70-second window finishes. Other errors are not billed. A 404 still costs credits, so a stale URL list quietly eats allowance, and a client timeout shorter than 70 seconds can get billed for pages you never received.
For a developer who wants proxies, retries and anti-bot handling behind one endpoint, yes: the Startup plan buys a million plain pages for $149. It is less of a fit for a business team without a developer, because the parsing, scheduling, storage and alerts around the endpoint are still code someone has to write and maintain.
Rows in your sheet instead of pages in your code
Describe the data, pick the schedule, and get named columns in Google Sheets, Excel or the API. Flat monthly plans from $79.