Stillindexed

Know before Google does. SEO disaster alerts for clients.

Alarms

SEO

Business Intelligence

Visit WebsiteSee on Product Hunt

I built this because an accidental noindex sits live for 24 to 72 hours before anyone notices.

Search Console tells you after the indexing damage is done, uptime monitors think a noindexed

page is fine because it loads, and audit suites crawl weekly, which is slower than a deploy.

It fetches the directives Google's indexing decision trusts first: robots.txt, meta robots,

X-Robots-Tag, canonical, status codes, redirect chains, title and H1, sitemap, TLS and domain

expiry - every 15 or 30 minutes per URL, diffs each fetch against the last, suppresses flapping,

then alerts by Telegram and email with what changed, where, when, and the before and after.

Public URLs only: no Google account access, no credentials, no snippet on the client's site.

HTTP only, no JavaScript rendering.

The part I want you to poke at is the proof. This product sells silence, and a working month

looks exactly like a dead monitor, so asking you to trust a dashboard would be worthless.

Instead there is a page that breaks itself on a published schedule. Two hours healthy, one hour

carrying noindex, one hour returning 503, keyed to the UTC hour and a log of every check the

monitor made against it, including the failures:

https://api.stillindexed.com/canary/log

Fetch the schedule, fetch the log, check whether they agree. Across 1,208 checks the median from

break to detection is 16 minutes, p90 29. If the log ever disagrees with the schedule, the

monitor is wrong and you should not buy it.

I also scanned the 10,000 most-visited homepages for the exact failures this watches. 87 carried

a meta noindex, 10 sent it as a header, 143 robots.txt files disallowed Googlebot at the root.

Most of that is deliberate or a bot wall, so I refetched all 225 flagged domains and classified

them by hand: 14 look like nobody meant it, including a national government portal, two US state

government sites and a university homepage. Method, limits and the full 10,000-row CSV are

published, and a scheduled job re-reads those 14 every week and publishes the result including

the rows it cannot classify.

Two honest limitations before you ask. There is no free trial. Every check is available as a

free ungated tool with no account instead, and paid plans carry an unconditional 30 day refund.

And I have no customers yet, so I have no testimonials, and you would be right not to believe

one.

11 free checkers, no signup: https://stillindexed.com/tools

Happy to run a check against any domain in the comments and post the exact line and the curl.