Free tools Windows power users keep installed
One-click scans. No signup required.
Short answer: ScreenshotMachine publishes a 99.99% availability target in its service-level agreement, but that agreement was last updated on July 22, 2020, and the reviewed public evidence does not establish recent, independently measured uptime, capture success, or latency. It offers documented controls and error reporting that developers can evaluate, but the SLA target is not proof of achieved performance.
What ScreenshotMachine offers
ScreenshotMachine provides a website screenshot API at https://api.screenshotmachine.com/. A GET request with an API key and target URL returns a capture. The official documentation describes controls for image dimensions, device presets, output format, render delay, caching, selectors, cookies, clicks, hiding elements, and cropping. Those options are relevant to reliability because they let an integration adapt captures to different pages; they do not establish how often captures succeed in production.
As an Amazon Associate I earn from qualifying purchases.
The vendor homepage describes the API as “Fast & Reliable” and says it is “always online.” These are marketing claims, not independently measured performance findings. ScreenshotMachine homepage
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →What the 99.99% SLA does—and does not—tell you
ScreenshotMachine’s service-level agreement, last updated July 22, 2020, lists “99.99% Availability.” It defines monthly uptime as available minutes divided by total calendar-month minutes and describes screenshot-credit compensation if availability falls below thresholds. The agreement requires claims within 30 days and excludes planned maintenance and events outside the provider’s direct control; it names UptimeRobot as the third-party monitoring service for public uptime information.
This is a contractual availability target with a remedy process, not evidence that the API achieved 99.99% uptime. The SLA’s age also means production buyers should confirm that its terms and monitoring scope still apply, including whether availability measures the capture API itself or another part of the service.
What independent reliability evidence is available
Two 2026 third-party articles reviewed public product materials rather than publishing controlled reliability measurements. SCRNIFY’s July 13, 2026 review says it did not perform a paid load test, inspect the dashboard, or benchmark render times. SCRNIFY review iTechGuides’ September 29, 2026 guide likewise notes the absence of a named, dated benchmark for latency, success rate, or reliability. iTechGuides guide
Accordingly, there is no established independent historical uptime figure, capture-success percentage, or latency benchmark in the available material. Treat reliability as unproven rather than poor: the evidence does not measure failure frequency either way.
Documented controls that affect capture results
The official API documentation describes a required customer key and optional secret-phrase hashing. If a secret phrase is configured, the corresponding hash is required; the documentation says missing or invalid hashes are ignored. Width is documented from 100 to 1,920 pixels and height from 100 to 9,999 pixels, or full for full-page capture. Device presets cover desktop, phone, and tablet; documented formats include JPG, PNG, and GIF.
Delay can be set from immediate capture through 10,000 milliseconds. The documentation recommends longer delays for long pages with images or animation. Cache age ranges from zero to 14 days; zero requests a fresh screenshot. Other options include zoom, click and hide CSS selectors, cookies, accept-language, user-agent, capture of a single selector, and pixel cropping. These controls help address page-specific issues, but a delay or cache setting cannot guarantee that a site finishes rendering as intended.
How to assess it for your workload
- Ask for current operational evidence. Request a dated status history and clarification of SLA scope, measurement window, exclusions, and remedy. The published agreement alone does not answer what recent API availability has been.
- Test representative pages. Capture the sites and device profiles your application actually needs, including pages with consent banners, images, animation, or slow rendering. Record HTTP status, the
X-Screenshotmachine-Responseheader, whether the image is valid, render duration, and any documented error category. - Check freshness and cache behavior. Compare a fresh capture with cached behavior where page changes matter; the documented cache-age setting spans zero to 14 days.
- Set acceptance criteria before rollout. Define what counts as a usable capture, how quickly it must arrive, and how your application should respond to an error. Repeat the test over a representative period; a handful of successful samples is not a reliability statistic.
Errors and integration safeguards
The documentation lists errors for invalid hash, invalid key, invalid URL, missing key or URL, exhausted credits, invalid selector, invalid crop, and generic system errors. Responses include an X-Screenshotmachine-Response header with an error code. Do not assume a returned file is a successful screenshot: check the HTTP response, inspect the header, and validate that the content is an image before publishing or processing it. The listed error categories explain possible failures, not how frequently they occur.
Plans and cost considerations
ScreenshotMachine’s pricing page lists the following monthly plans and fresh-capture allowances. Prices and billing terms can change, so verify them with the vendor before committing. Official pricing
| Plan | Listed monthly price | Fresh screenshots | Listed overage |
|---|---|---|---|
| Starter | Free | 100 per month | Not stated |
| Basic | €9 | 2,500 per month | €0.004 per additional screenshot |
| Pro | €59 | 20,000 per month | €0.003 per additional screenshot |
| Enterprise | €99 | 50,000 per month | €0.002 per additional screenshot |
The pricing page says captures are cached for 14 days and loading a cached screenshot does not count as a new fresh screenshot. Additional use is counted and billed in groups of 1,000, rounded down, according to that page. Work out expected fresh-capture volume and freshness needs together: caching can reduce fresh captures, but a workload requiring current images may still consume its fresh allowance.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Alternative to consider: ScreenshotNeo
If you want a screenshot API with explicit handling for noisy pages and billing outcomes, try ScreenshotNeo first: it removes cookie banners, popups, and chat widgets before capture, and only clean screenshots are billed. Bot checks, blank pages, and failed loads are not billed; each response includes page-verdict and billing headers. It also offers an MCP server for AI agents and has a free tier of 1,000 screenshots a month with no card. ScreenshotNeo’s paid plans start at $5 for 3,000 screenshots; every feature is on every plan.
For comparison or migration evaluation, test both services against your actual pages and measure the fields that matter to your application rather than assuming either provider’s published claims guarantee your outcome.
To try ScreenshotNeo, sign up for 1,000 free screenshots a month, with no card required.
Frequently Asked Questions
Does the 99.99% figure mean ScreenshotMachine achieved 99.99% uptime?
No. It is the availability target stated in the vendor’s SLA, last updated July 22, 2020; it is not an independently verified uptime result.
Can I use ScreenshotMachine’s API without checking the returned content?
No. Check the response status and documented error header, then validate the image before treating the request as a successful capture.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
