Home / Terms / Successful request
What is a successful request?
A successful request is one the vendor counts as delivered, and so bills. That is not the same as getting the data you wanted: ScraperAPI and ScrapingBee bill a 404 page (ScrapingBee a 410 as well), and Firecrawl bills a 403 or a 404, while most APIs do not bill an attempt that was blocked or timed out. The table below gives each API’s rule, with the date we read it. (ScraperAPI credit rules, checked 2026-10-08; ScrapingBee FAQ, checked 2026-10-07; Firecrawl pricing, checked 2026-10-08.)
Successful request across the APIs we track
| API | Failed requests | What counts as success | Checked |
|---|---|---|---|
| Brave Search API | Bills successful requests only | Failed requests do not count against the request quota. | Oct 9, 2026 |
| Bright Data SERP API | Bills successful requests only | By default only successful responses are billed, and failed or errored requests and internal retries are not billed. If custom headers or cookies are selected for the zone, all requests are billed regardless of outcome. | Oct 9, 2026 |
| Bright Data Web Unlocker and Web Scraper API | Bills successful requests only | CPM is the cost of 1,000 successful requests, so only successful Web Unlocker API requests count toward billing. The exception is Custom Web Unlocker API features such as manual headers, cookies, or expect elements: with any of those enabled, 100% of requests are billed, successful and failed. | Oct 9, 2026 |
| Firecrawl | Bills failed requests too | A page that responds with an error status such as 403 or 404 is still returned to you and costs 1 credit. | Oct 7, 2026 |
| HasData | Bills successful requests only | Credits are deducted only for successful results; failed or blocked requests do not cost credits. | Oct 8, 2026 |
| Kernel | Bills failed requests too | Kernel charges active browser resource use rather than successful page results, so active compute consumed before a failure remains billable. | Oct 9, 2026 |
| Linkup | Bills successful requests only | No credit is deducted for errors, missing parameters, internal errors, or a response with no relevant result. | Oct 9, 2026 |
| Nimble | Bills successful requests only | The self-serve table prices successful URLs. Failed retrieval attempts are not included in the stated per-successful-URL charge. | Oct 9, 2026 |
| Olostep | Bills successful requests only | No, Olostep does not charge for failed requests. You are only billed for successful requests, ensuring predictable and fair usage-based pricing. | Oct 9, 2026 |
| Oxylabs SERP Scraper API | Bills successful requests only | Failed scraping attempts due to our system errors (5xx and 6xx status codes) don’t count, and you won't be billed for them. | Oct 7, 2026 |
| Oxylabs Web Scraper API | Bills successful requests only | Failed scraping attempts due to our system errors (5xx and 6xx status codes) don’t count, and you won't be billed for them. | Oct 7, 2026 |
| Scrape.do | Bills successful requests only | Only 2xx, 400, 404 and 410 responses consume credits. Other failed or blocked responses are not charged, and an invalid HTTP 200 result can be submitted for review. | Oct 9, 2026 |
| ScrapeOps | Bills successful requests only | Every successful request to the ScrapeOps Proxy API Aggregator that returns a successful status code 200 or 404 consumes API credits from your monthly API credit quota. | Oct 9, 2026 |
| ScrapeOwl | Bills successful requests only | HTTP 200 and HTTP 404 responses consume credits; the documented 400, 401, 403, 429, and 500 responses do not. | Oct 9, 2026 |
| ScraperAPI | Bills successful requests only | Only HTTP 200 and 404 responses are charged. A client-cancelled request is also charged if cancelled before ScraperAPI has had 70 seconds to finish. | Oct 9, 2026 |
| Scraping Robot | Bills successful requests only | Only a valid result returned without errors is charged; failed results and timeouts are not charged. | Oct 9, 2026 |
| ScrapingAnt | Bills successful requests only | Blocked and error responses are not billed. ScrapingAnt describes billable delivery as HTTP 200 with content. | Oct 9, 2026 |
| ScrapingBee | Bills successful requests only | We only charge for successful requests, i.e returning with a 200, 404 or 410 status code. | Oct 7, 2026 |
| ScrapingBot | Bills successful requests only | Refunded automatically: timeouts, blocks and other 4xx answers such as 401, 403 and 429, any 5xx from the target, and errors on our side. | Oct 9, 2026 |
| ScrapingDog | Bills successful requests only | Failed requests never charged | Oct 9, 2026 |
| Scrappa | Bills successful requests only | Failed API requests do not consume credits. | Oct 9, 2026 |
| SearchApi.io | Bills successful requests only | Only responses with HTTP status 200 consume searches; failed requests are not charged. | Oct 9, 2026 |
| SerpApi | Bills successful requests only | Only successful searches are counted toward your monthly searches. Cached, errored, and failed searches are not. The number of results returned per response will not affect the number of credits used—responses with 100 results or empty result sets will both count as 1 search. | Oct 5, 2026 |
| Serper.dev | Bills successful requests only | One credit is deducted for each successful response. | Oct 9, 2026 |
| SERPHouse | Bills successful requests only | The vendor states that only successful requests are charged. | Oct 9, 2026 |
| serpstack | Bills successful requests only | API errors are not counted toward the monthly quota. | Oct 9, 2026 |
| TinyFish | Bills successful requests only | Search and Fetch never draw from the wallet, so a failed Search or Fetch call cannot produce a charge. Failure billing for the paid Agent and Browser APIs is not published. | Oct 9, 2026 |
| WebScraping.AI | Bills successful requests only | Failed requests do not consume credits. | Oct 9, 2026 |
| WebScrapingAPI | Bills successful requests only | The pricing page states that only successful requests are charged. A complete list of billable HTTP status codes was not published. | Oct 9, 2026 |
| ZenRows | Bills successful requests only | You only spend credits on successful requests. Failed attempts aren't charged, and 404 or 410 responses count as usable results. | Oct 9, 2026 |
45 more APIs do not publish whether a failed request is billed; this site prices them as if it is, and says so on each bill.
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.
The definition changes a bill whenever a workload contains missing pages, access-denied pages, target-side errors, cancellations, or empty but technically delivered responses. Counting only useful rows in a forecast can understate billable volume if the vendor counts more HTTP outcomes. This is only an accounting distinction. (ScraperAPI credit rules, Firecrawl pricing, and Bright Data Web Unlocker billing, checked 2026-07-26.)
ScraperAPI charges responses with HTTP 200 or 404. It also says a client-cancelled request is charged when cancellation occurs before the service has had seventy seconds to finish. Other response codes are not charged under its published credit rule. (ScraperAPI credit rules, checked 2026-07-26.)
ScrapingBee charges HTTP 200, 404, and 410 responses. That means a known missing or gone page can consume credits even though it produces no current content to extract. Other failed responses do not consume credits under the published FAQ rule. (ScrapingBee FAQ, checked 2026-07-26.)
Firecrawl separates service-side failure from a response delivered by the target. It does not charge for Firecrawl-side timeouts or server errors, but it does charge when it fetches a page and the target returns a 4xx or 5xx response. A target error can therefore be billable even though the requested content was not obtained. (Firecrawl pricing, checked 2026-07-26.)
Bright Data Web Unlocker says its request charge applies only to successful deliveries and that failed deliveries are not charged. Its reviewed billing wording does not turn that rule into a promise that every delivered response contains the buyer’s desired data. (Bright Data Web Unlocker billing, checked 2026-07-26.)
Before signing up, ask for the exact billable status list, cancellation rule, timeout rule, cache treatment, and process for disputing a technically successful but unusable body. In a trial, log the HTTP status, vendor status, response body check, and credit debit for the same request. Forecast both billable responses and useful outputs as separate counts. (ScraperAPI credit rules, ScrapingBee FAQ, Firecrawl pricing, and Bright Data Web Unlocker billing, checked 2026-07-26.)