Status & reliability

ScrapeIt publishes live, platform-wide reliability numbers for every active product on a public page — no account needed:

app.scrapeit.io/status

The page shows an overall status, the success rate and response time of each product for the last hour, last 24 hours and last 7 days, and a 30-day daily success-rate chart. It refreshes every minute.

How the metrics are defined

Metric	Definition
Success rate	Requests that delivered the target website's own response ÷ requests the platform executed.
p50 / p95	Median and 95th percentile of the end-to-end response time of executed requests, in milliseconds.
Executed request	A request that passed validation and limits and was actually run by the platform. Shared-cache hits are not counted (they would flatter both metrics).

What counts as a success: the target answered and the result reached you — including the website's own HTTP 4xx/5xx and empty pages, because that is the website's real answer.

What counts as a failure: proxy, network, browser, internal and timeout failures, and anti-bot block or challenge pages we could not get past.

These definitions are for the success rate only. What a request costs is on Billing & credits.

What is not counted at all: requests refused before execution because of the request itself or your plan — VALIDATION_ERROR, an invalid API key, insufficient credits, CONCURRENCY_LIMIT_EXCEEDED and URLs blocked by the private-address (SSRF) policy. These say nothing about whether the platform works.

Thresholds and privacy

• A window or day is published only with at least 50 executed requests from at least 3 different customers. Below that the page reads "Not enough data" — a small sample would be noise, and a product used by one or two customers would otherwise expose exactly their traffic.
• No request volumes, customer identifiers or target domains are ever published. Rates are rounded to 0.1%.
• A product is degraded when its success rate in the last hour — or the last 24 hours, if the last hour is too quiet — is below 90%.
• The overall status is Service outage when the API cannot reach its own databases, Degraded performance when any product is degraded or the metrics are temporarily unavailable, and All systems operational otherwise.

Status API

The same data is available as JSON, without authentication:

curl "https://api.scrapeit.io/api/v1/public/status"

{
  "data": {
    "status": "operational",
    "generated_at": "2026-10-22T12:00:00.000Z",
    "components": { "api": "operational", "metrics": "operational" },
    "products": [
      {
        "product_code": "web_scraping_api",
        "product_name": "Web Scraping API",
        "status": "operational",
        "windows": {
          "last_hour": {
            "sufficient_data": true,
            "success_rate": 0.987,
            "p50_ms": 850,
            "p95_ms": 5640
          },
          "last_24_hours": {
            "sufficient_data": true,
            "success_rate": 0.975,
            "p50_ms": 900,
            "p95_ms": 6100
          },
          "last_7_days": {
            "sufficient_data": false,
            "success_rate": null,
            "p50_ms": null,
            "p95_ms": null
          }
        },
        "daily": [{ "date": "2026-10-22", "success_rate": 0.97 }]
      }
    ],
    "methodology": { "min_requests": 50, "min_customers": 3, "degraded_below_success_rate": 0.9 }
  }
}

The snapshot is cached for up to a minute, so polling faster returns the same data. The endpoint shares the per-IP request rate limit and answers 429 RATE_LIMIT_EXCEEDED with Retry-After when exceeded.