We break a page on purpose so you can watch the monitor catch it
stillindexed has no customers yet, so it has no testimonials, and you would be right not to believe one. What we can offer instead is a page you can fetch yourself, breaking on a schedule published in advance, and a log of what the monitor did about it.
The schedule
api.stillindexed.com/canary changes state on the clock. Nothing is stored and there is no switch to flip: the state is worked out from the current UTC hour every time the page is requested, which is why the schedule can be published ahead of time.
| UTC hour mod 4 | State | What it serves |
|---|---|---|
| 0 | healthy | 200, indexable |
| 1 | healthy | 200, indexable |
| 2 | noindex | 200 carrying a noindex tag |
| 3 | error | 503 |
UTC hour mod 4: 0 and 1 healthy, 2 noindex, 3 error (503). One full cycle every 4 hours.
Check it yourself, right now
Work out what the hour should be. Fetch the page. Then read the log and see whether it agrees. The log shows every check including the ones that failed, every incident including the ones the flap pass decided were noise, and which channel each alert actually reached.
During its error hour the page answers 503 on purpose, so a browser will show you an error. That is the point of it.
What this proves, and what it does not
It proves the pipeline runs on a schedule, that a change is detected and confirmed rather than fired on first sight, that flapping is suppressed, and that an alert reaches a channel. Those are the parts that quietly stop working in a monitor, and the parts you cannot verify from a marketing page.
It does not prove we will behave correctly on your stack. A scheduled break is a rehearsal, and the honest answer to that is not to argue about it: every check this product runs is also a free tool with no account, so you can point one at a real client URL and judge the output yourself.
The same engine ran a one-pass scan of the 10,000 most-visited homepages: the homepage indexability study lists every robot that failed to serve a sitemap it declared. If the log ever disagrees with the schedule above, the monitor is wrong and you should not buy it. That is the bet, and it is the reason the log publishes the failures alongside the successes.
Because the schedule is exact, the log can also state how long each step took: thedetection latency block on the canary log shows the median and p90 from the scheduled change to detection, to confirmation, and to the first alert delivered, measured on a 30-minute cadence — a paid Agency site is checked every 15 minutes, so its numbers are shorter, and we publish the slower one.
What you get when it is your pages
The same pipeline, pointed at the URLs you care about. Starter is $39 a month for 15 sites, checked every 30 minutes, with site level checks daily. Card first, no trial, and a 30 day refund if it is not what you wanted.
Choose Starter