Scrapfly Pricing and Scrapfly Cost per 1,000 Pages on Every Plan

Scrapfly is a scraping API with a good reputation among developers: managed proxies, browser rendering, an anti bot layer it now calls the Unblocker, plus screenshot, extraction and crawler endpoints on the same key. It also writes some of the best scraping tutorials on the web, which is how many teams first meet it.

Its pricing is a credit system, and credit systems are easy to misread. The plan figures below come from scrapfly.io/pricing and the Scrapfly billing documentation, read on October 8, 2026. Prices are monthly in USD. Check the live page before you commit.

The Scrapfly plans and what each includes

Annual billing gives two months free, about 16 percent off. Pro and above get priority email support. Scrapfly publishes no uptime or response time SLA on self serve plans; its Unblocker docs put an SLA plan at a minimum commitment of 50,000 USD a month. SOC 2 Type II and ISO 27001 are listed for every customer, which matters if your security team reviews vendors. If that review is a spreadsheet passed around by email today, a compliance management tool that maps each vendor's SOC 2 report to your controls saves the next audit a week.

What one Scrapfly request costs in credits

Every request starts with a proxy pool. Datacenter is the default at 1 credit. Residential is 25 credits. Browser rendering with render_js=true adds 5. So the four combinations are 1, 6, 25 and 30 credits. Binary downloads such as PDFs and images bill by size instead, after the first free megabyte: 3 credits per 100 KB on datacenter, 10 on residential.

Two smaller rules move the bill. A cached response within your TTL costs 1 credit instead of the full price, which helps if you revisit the same pages often. And failed scrapes are protected, but only while failures stay under 30 percent of eligible traffic in any hour. Cross that line and Scrapfly bills the failures too, which is exactly when a target has tightened its defenses and your error rate is climbing.

Scrapfly cost per 1,000 pages

Divide the plan price by its credits and multiply by the credits per page. These are included credit rates; overage on Pro is 3.50 USD per 10,000 credits, which is 3.5 times the included rate.

Startup lands on the same per credit rate as Pro, 0.10 USD per 1,000 credits, and buys you more concurrency and projects rather than cheaper pages. Our plans price every successful page the same: 0.98 USD per 1,000 on Starter, 0.60 on Growth and 0.33 on Scale, residential proxies and rendering included.

That gives a clean breakeven. Growth against Pro ties at exactly 6 credits a page, which is a datacenter page with a browser. Starter against Discovery ties at about 6.5 credits. Below that line Scrapfly is cheaper. Above it, which means any page that needs residential traffic, a flat price per page is cheaper by about four to eight times.

The Unblocker line nobody can price in advance

Protected sites are where a scraping budget is actually spent, and Scrapfly's docs are candid about it. The Unblocker may switch your request to a better proxy pool or turn on the browser on its own, and those changes raise the credit price. On top of that, specific targets carry fees that, in Scrapfly's words, "are not publicly documented" and show up only in the cost section of your logs. The docs add that the credit cost may change when a site adds protection or moves to another anti bot vendor.

So for the pages that matter most, the price per page is something you measure after the fact. Scrapfly gives you tools for it: a cost_budget parameter that fails a call above a set price, per project and per domain limits, and an X-Scrapfly-Api-Cost header on every response. Use them. Run a few hundred requests against each real target, read the header and only then convert your volume into a plan. Our Cloudflare scraping API and DataDome bypass API pages explain what those protected requests involve on our side, at the same flat page price.

Real monthly budgets on Scrapfly and on flat per page pricing

Three workloads priced on the cheapest Scrapfly plan that fits, using list rates and the published credit costs, before any target specific Unblocker fees.

One detail decides the second row. Pro with overage would come to 275 USD, but pay as you go is capped by default at 50 percent of your monthly quota, and 1.5 million credits on Pro sits exactly on that cap. One busy week and the integration stops. Startup is the plan that actually holds that workload. Discovery is worse still for protected work: its 200,000 credits are a hard cap, about 6,600 residential pages with a browser, and then nothing until next month.

Where Scrapfly is the right buy

Plain pages at volume. Public sites that do not block datacenter traffic cost 0.09 to 0.15 USD per 1,000 pages on Scrapfly, cheaper than our plans. If most of your list is documentation, government data or small business sites, the credit math works in your favor.

Teams that want screenshots, extraction and a cloud browser on one balance. Scrapfly's credits cover all of them on one key, and its Python and TypeScript SDKs are well kept. Teams that already learned scraping from its blog often find its API the shortest path.

Small tests. The 1,000 free credits and the 30 USD Discovery plan are a cheap way to try a target, as long as the target is not heavily protected.

Where flat per page pricing wins

Protected targets. Retail catalogs, marketplaces, travel, ticketing and job boards run Cloudflare, DataDome, Akamai or Imperva and need residential traffic on most requests. At 25 to 30 credits each, that is where Scrapfly's bill grows, and where the Unblocker can add fees you cannot see in advance. One price per successful page removes both problems. Review sites belong here too: our Trustpilot scraper API reads company pages that answer plain clients with a WAF block.

Budgets that finance has to sign. 149 USD for 250,000 pages is a number a controller can approve. A credit meter with a 30 percent failure clause and undocumented per target fees is harder to forecast.

If you are comparing other credit based vendors too, the same per 1,000 page arithmetic for ScrapingBee, ZenRows and ScraperAPI is in our pricing series, and every plan of ours is on the pricing page.

Questions buyers ask

How much does Scrapfly cost? Discovery is 30 USD a month for 200,000 credits, Pro 100 USD for 1,000,000, Startup 250 USD for 2,500,000 and Enterprise 500 USD for 5,500,000. Custom contracts start at 1,200 USD a month. Annual billing takes about 16 percent off every plan.

Does Scrapfly have a free plan? New accounts get 1,000 API credits with no credit card, capped at 5 concurrent requests and with no overage. That covers about 1,000 plain pages or 33 protected pages with a browser, enough to test an integration but not to run a scraper.

How many credits does a Scrapfly request use? 1 credit on datacenter proxies, 25 on residential, plus 5 when browser rendering is on, so 1, 6, 25 or 30. The Unblocker can raise that by upgrading the proxy or turning on the browser, and some targets carry extra fees shown only in your logs.

What happens when Scrapfly credits run out? Discovery and Free stop at the quota. Pro and above continue on pay as you go at 3.50, 2.00 or 1.20 USD per 10,000 credits, capped by default at 50 percent of your monthly quota unless support raises it.

Does Scrapfly charge for failed requests? Not normally. Failed scrapes with eligible status codes are free, but if more than 30 percent of eligible traffic fails within an hour, the protection is switched off and the failures are billed. Codes such as 404 and 410 are excluded from protection and always billed.

Is Scrapfly cheaper than a scraping API with flat pricing? For plain datacenter pages, yes. For pages that need residential proxies and a browser, no: Scrapfly works out to 2.73 to 4.50 USD per 1,000 before Unblocker fees, against 0.33 to 0.98 USD per 1,000 on our plans.