A dev ships a change on a Tuesday. Nobody looks at it again. Six weeks later someone asks why ChatGPT stopped citing your docs. We re-audit your tracked pages on a schedule so it surfaces in days, not at the next quarterly review.
CompareTwo runs of the same URL, held against each other. The blocked crawlers and the speed drop both landed between 14 July and 11 August.
This is where you find out what moved. Choose the run from before the deploy and the run from after, and every measured value sits next to its earlier self.
Reading itTwo crawlers newly refused and the server got slower, while the markup did not move. That points at infrastructure — an edge rule or a WAF change — rather than at anything the content team did.
We would rather be exact about this than sell you a promise the product does not keep yet.
Compared against the previous run.
Compared against the previous run.
Compared against the previous run.
Compared against the previous run.
Every category above carries its own dial, because a blocked crawler and a reworded subheading do not deserve the same interruption.
Everything, including things that got better.
Edits a person made, and grades that moved the wrong way.
The quiet setting, for pages you want to hear about once.
Star a page to track it. Up to 100 URLs, and it is worth choosing the commercially important ones rather than everything.
Each project gets a fixed hour of the day, spread so runs do not all land at once. Change it under Time (UTC) if you would rather have results waiting at the start of your day.
An alert points at the run. Open it, or open Compare Mode against an earlier run to see the movement in full.
| Plus | Pro | Agency | |
|---|---|---|---|
| Scheduled monitoringAutomatic re-audits of tracked pages. | — | — | ✓ |
| Tracked URLsPages re-audited on the schedule. | — | — | 100 |
| Compare ModeAny two runs of the same page, side by side. | — | — | ✓ |
| Email alertsSent when a tracked page moves. | — | — | ✓ |
Track the pages that earn, let them re-audit on a schedule, and compare any two runs when something moves.
It re-audits the pages you have chosen to track, on a schedule, and emails you only when something differs from the previous run. A run where nothing changed produces no email — it is there so you find out about a regression before a client does.
The Agency plan only. Tracked URLs are an Agency feature, and scheduled runs only happen on tracked URLs, so no runs execute on Plus or Pro.
Daily or weekly, at a fixed hour. Each project is given an hour so runs are spread across the day rather than all landing together; set it to whatever suits you under Time (UTC).
Four, and you choose which. Crawl access — robots.txt, HTTP status, noindex, bot access. Speed and stability — a TTFB, CLS or INP grade moving. Authority — structured data and check results changing. Content — what a crawler can read on the page changing. Leave a category off and it will not alert you.
Yes, for all four — you get the previous value and the new one. Content names what kind of thing moved: the headings, the main heading, or the body text. It does not quote the exact wording that changed; Compare Mode is where you see that. We would rather state the limit than have you discover it.
Yes. Each of the four categories has its own sensitivity: any change, meaningful changes, or only significant changes. Content and speed start on meaningful changes, so a moved date stamp or a grade that improved will not interrupt you. Crawl access and authority start on any change, because a blocked crawler earns an email on its own. Lower the noisy category rather than switching it off — off means you will not hear about it when it genuinely breaks.
Four things to check, in order. The change category has to be enabled — a change in a category you turned off does not alert. The sensitivity for that category has to be low enough to count it — on “only significant changes” a small edit is ignored on purpose. The “changed pages” notification has to be on. And alerts come from contact@beseenby.ai, which is worth looking for in spam once.
Fewer than you can. A hundred tracked pages produce a hundred pages of noise; the pages that earn money produce alerts you will act on.
Yes. Pausing keeps the schedule and the alert configuration and simply skips runs until you turn it back on. Untracking a page removes it from future runs on its own.
Any two runs of the same page still inside your report history, which is a year on Agency — not just consecutive ones. That is what makes “before the migration versus now” possible months later.
Yes. Each run counts as an audit against the same monthly pool as single, batch and site audits.