If you sell uptime, your own uptime is part of the spec sheet
Feature tables do not separate monitoring vendors from each other. They all ping things and tell you when the ping fails.
What separates them is whether they go down.
So the cat went round nine status pages and tried to pull each company's own incident history in a form something other than a pair of eyes could read.
One of the nine handed it over.
The nine
| Company | Incident history |
|---|---|
| Better Stack | RSS. 44 entries, reaching back to February 2023 |
| UptimeRobot | A feed exists. Its own robots.txt blocks it — see below |
| Oh Dear | RSS exists but returns nothing. The page itself shows only the last seven days |
| Cronitor | Rendered by JavaScript only |
| Healthchecks.io | No machine-readable feed |
| Checkly | Same |
| Pingdom | Same |
| StatusCake | Same |
| updown.io | The API returns 400 |
Every one of them has a status page a person can read. Nothing here is hidden.
What is missing is a form you can compare.
One company's robots.txt blocks its own feed
UptimeRobot publishes an incident feed. It lives here:
https://status.uptimerobot.com/api/getEventFeed/…The robots.txt on that same host says this:
User-agent: *
Disallow: /api/
The feed is under /api/. The company's own robots.txt closes the door on the company's own feed.
This reads like one line written to protect an admin API, catching a public feed on the way past. It looks like an accident rather than a decision. But to anything that honours robots.txt, an unreachable feed and a missing feed are the same feed.
Cronitor runs its status page on a competitor
One other thing turned up.
Open status.cronitor.io and the page is served by Hyperping — a company that sells monitoring and status pages, in the same market as Cronitor.
There are good reasons not to host your own status page on your own product. The thing that watches and the thing being watched should not share a failure. A vendor that detects its own outage with its own monitoring and reports it on its own status page has built something that goes dark all at once.
Still, a monitoring company buying monitoring from a competitor is a fact worth knowing.
Do not compare the counts
The tempting mistake here is to line up the number of incidents and read it as a ranking.
Feeds cap. Among the companies this site tracks outside monitoring, Linear and Notion both stopped at exactly 25 entries — sitting on a limit. Better Stack returned 44 and Plausible 49, neither declaring a cap.
A company you can see 25 incidents from does not have fewer incidents than one you can see three years of. You can just see less of it.
So every record here notes whether the feed was cut off by a count limit or by a time window. Without that, the companies that disclose least look best.
Prices can be recovered. Uptime cannot
Pricing pages sit in the Internet Archive going back years. A price rise from 2024 can still be reconstructed today, which is most of what this site does.
Incident feeds are not like that.
Neither status.betterstack.com/feed.rss nor status.plausible.io/feed.rss has a single capture in the archive. Nobody saved them. Better Stack's three and a half years did not come from digging through the past — it came from asking once, today, and getting an answer.
Which means an incident history that is not published now cannot be recovered later. The day a company folds its feed, its past goes with it.
That is why the cat started saving these daily rather than planning to.
What could not be established
Severity and duration are not compared here. Counts and dates came out of the feeds reliably; severity labels did not, because each company words them differently. How long you were down matters more than how often, so this remains the gap.
For the companies with no machine-readable feed, the human-readable status pages were not read. Reading a page tells you something once. Only a feed lets you keep comparing. This article is about the second thing.
Sources
Each company's official status page and its robots.txt were fetched directly in August 2026.
- Better Stack Status
- UptimeRobot Status / robots.txt
- Oh Dear Status
- Cronitor Status
- Healthchecks.io Status
- Checkly Status
- Pingdom Status
- StatusCake Status
- updown.io Status
Only what each vendor publishes itself is used. No user reviews or forum posts.