Home / Terms / Asynchronous scraping
What is asynchronous scraping?
Asynchronous scraping means sending a job and collecting the result later, instead of waiting on an open connection for the page to come back. The API answers at once with a job ID; your code then asks for the result when it is ready (polling), or the API sends it to an address you gave it (a webhook). Synchronous scraping, the default on most APIs, holds the request open until the page is fetched, rendered and returned.
Asynchronous scraping across the APIs we track
| API | In the same response | Submit, then collect | Webhook | Batches of URLs | Checked |
|---|---|---|---|---|---|
| Anchor Browser | Yes | Yes | Yes | not published | Oct 12, 2026 |
| Apify | Yes | Yes | Yes | not published | Oct 9, 2026 |
| Bright Data SERP API | Yes | not published | Yes | Yes | Oct 12, 2026 |
| Bright Data Web Unlocker and Web Scraper API | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Browsable | Yes | Yes | not published | not published | Oct 12, 2026 |
| Browse AI | not published | not published | Yes | Yes | Oct 12, 2026 |
| Browser Use Cloud | not published | Yes | Yes | not published | Oct 12, 2026 |
| Browserbase | Yes | Yes | Yes | not published | Oct 12, 2026 |
| Browserless | Yes | Yes | Yes | not published | Oct 12, 2026 |
| Cloudflare Browser Rendering | Yes | Yes | not published | not published | Oct 12, 2026 |
| Crawlbase | Yes | Yes | Yes | not published | Oct 12, 2026 |
| DataForSEO SERP API | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Decodo Web Scraping API | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Diffbot | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Exa | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Firecrawl | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| HasData | Yes | Yes | not published | Yes | Oct 12, 2026 |
| Hyperbrowser | Yes | Yes | not published | Yes | Oct 12, 2026 |
| Kadoa | not published | Yes | Yes | Yes | Sep 2, 2026 |
| Kagi Search API | Yes | not published | not published | Yes | Oct 12, 2026 |
| Linkup | Yes | Yes | not published | Yes | Oct 12, 2026 |
| Nimble | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Olostep | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Oxylabs SERP Scraper API | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Oxylabs Web Scraper API | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Parallel Web Systems Search API | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Perplexity Search API | Yes | not published | not published | Yes | Oct 12, 2026 |
| Scale SERP | Yes | Yes | Yes | Yes | Oct 9, 2026 |
| Scrape.do | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| ScrapeGraphAI | Yes | Yes | Yes | not published | Oct 12, 2026 |
| Scrapeless | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| ScraperAPI | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Scrapfly | Yes | not published | Yes | Yes | Oct 12, 2026 |
| Scraping Robot | Yes | not published | No | not published | Oct 12, 2026 |
| ScrapingDog | Yes | not published | Yes | not published | Oct 12, 2026 |
| Scrappa | Yes | not published | Yes | not published | Oct 12, 2026 |
| Search1API | Yes | Yes | not published | Yes | Oct 12, 2026 |
| SerpApi | Yes | Yes | not published | not published | Oct 12, 2026 |
| SERPHouse | Yes | Yes | not published | not published | Oct 12, 2026 |
| SerpWow | Yes | Yes | Yes | Yes | Oct 9, 2026 |
| Spider.cloud | Yes | not published | Yes | not published | Oct 12, 2026 |
| Tavily | Yes | Yes | not published | Yes | Oct 12, 2026 |
| TinyFish | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| ValueSERP | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Valyu | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| WebScrapingAPI | Yes | Yes | not published | not published | Oct 12, 2026 |
| You.com API | Yes | Yes | not published | Yes | Oct 12, 2026 |
| ZenRows | Yes | Yes | Yes | Yes | Oct 12, 2026 |
| Zenserp | Yes | not published | Yes | Yes | Oct 12, 2026 |
22 more APIs publish only the same-response mode. "Not published" means the vendor's documentation does not mention the mode, which is not proof it is missing.
In more detail
The vendor examples below were read on the dates given in the text; the table above holds each vendor's current figure.
When it matters
A synchronous request is the simplest to write and suits a few pages at a time. It becomes awkward when pages are slow: a rendered page behind bot protection can take 30 seconds or more, and every one of those seconds holds a connection and one of your plan’s concurrent slots. Asynchronous delivery lets a scraper submit thousands of URLs in a batch, move on and collect results as they finish, which is how large crawls and scheduled jobs are usually built.
Does it change the price?
On most APIs the delivery mode does not change what a page costs: the same credits are charged whether the result comes back in the response or by webhook. Some SERP APIs are the exception, selling a queued mode for less than live results; the SERP API page prices both. The table shows which delivery modes each vendor documents.