Summary of an article published on The Web Scraping Club on 2026-09-01 (free to read). Read the full article
AntsData is a data provider you may not know. It sells pre-compiled datasets, SERP APIs and Web Search APIs, with a Web Unlocker API listed as coming soon. For anyone who needs Google results or Amazon data without running proxies, solving CAPTCHAs or building custom infrastructure, the practical questions are how easy the APIs are to call and what each run costs.
We tested two of the services. First the Google SERP API, a synchronous endpoint, which I ran from the playground and then moved into a Python script using requests. Then the Amazon Web Search API, which is asynchronous, pointed at Best Sellers category URLs. For both I checked the billing page to see what each run cost. I close with the pros and cons from these tests, including how the documentation, the playground and the MCP integration held up.
What’s inside
- The difference between the synchronous SERP endpoints and the async submit, poll, fetch model, and which kind of job fits each.
- Which Search endpoint parameters we set for the Google query: results type, location, language, Google domain, page and result limit.
- How the Amazon Best Sellers discovery by category URLs works, and what the JSON returned for each category contains.
- A check of the MCP integration that AntsData advertises for its Web Scraper APIs, and how the catalog compares with competitors of similar size.
In the full article: The Python script built from the playground code, the parameter screens, the JSON samples from both APIs, the billing figures per run, and the full pros and cons list.