Scrape.do sells a general web scraping API and, more interestingly, a set of finished scrapers for specific named targets.

Google Search, Google Maps, Google Shopping, Google Flights, Amazon, Walmart, YouTube, and even ChatGPT and Gemini.

Why Target Specific Endpoints Are Worth Paying For

A general scraping API gets you the page. You still have to work out what the page means.

For a complex target that is a substantial job. Google Flights is not a document, it is an application. Amazon product pages vary by category, region and seller. Google Maps results are assembled dynamically.

Writing parsers for those is difficult. Keeping them working is worse, because these are exactly the sites that change most often and give no notice.

A dedicated endpoint means somebody else absorbs that. You request a target and get structured fields, and when the site redesigns, their engineers fix it rather than yours.

The ChatGPT and Gemini Endpoints

Those two are unusual enough to mention separately.

Being able to query AI services programmatically through the same interface as everything else is a genuinely modern requirement, and it reflects how many pipelines now mix web data with model output.

Worth reading each service's own terms before building on it, as you would with any target.

WAF Bypassing Named Directly

Scrape.do talks about getting past web application firewalls, which is more specific than the usual language about avoiding blocks.

A WAF is a defensive layer sitting in front of a site, inspecting requests for patterns it considers automated. It is a different problem from IP reputation, and a clean address does not solve it on its own.

Naming it means acknowledging that the address is only part of the problem, which is more honest than most marketing manages. Our scraping guide covers the other layers.

A Free Open Source Client

There is a local first API client, free and open source, that keeps your data on your own machine.

That is a thoughtful thing to give away. Testing API requests usually means pasting data into somebody else's web tool, which is fine until the data is sensitive.

Proxies and a Scraping Browser Are Listed as Coming

Both appear in the product menu marked as not yet available, and flagging that rather than quietly listing them is worth crediting.

It also tells you the direction. A company that starts with finished scrapers and later adds raw addresses and a browser is building downwards through the stack, towards customers who want more control.

Worth checking current availability directly rather than planning around anything marked as coming, including on the strength of this paragraph.

Who It Fits

Scrape.do suits anyone scraping Google properties or major retailers, teams whose parsers keep breaking on complex targets, pipelines mixing web data with AI service output, and users facing WAF protected sites.

Who Might Look Elsewhere

If you need raw IP addresses for your own stack, a plain proxy provider is far cheaper.

Pricing here starts at the higher end of this market, so small or occasional users should compare carefully against per request alternatives.

Final Verdict

Scrape.do focuses on the targets that cause the most maintenance pain and sells the solved version.

A general scraping API, finished endpoints for Google properties, major retailers and AI services, WAF handling named explicitly, and a free open source client for testing.

Check whether your target already has a dedicated endpoint before writing anything. If it does, that is almost certainly cheaper than building and maintaining your own.