Root causes, not a spreadsheet
Four hundred missing descriptions is one template.
Every SEO tool is generous with findings and silent on priority. You get a list of 847 problems, sorted by severity in the abstract, and no view on which of them is worth an evening. So you fix the easy ones, or the loud ones, or none of them.
Missing descriptions across 400 blog posts is one templating bug, not 400 tickets. Grouped by issue type and URL pattern, with the confidence stated: 400 of 400 is a template, 6 of 14 is a question.
Then the ledger
A verdict, including the ones that say nothing moved.
Nothing to mark as done. We keep watching the affected pages and date the fix to when the defect actually disappeared from all of them. Comparable URLs are frozen at that moment, so a seasonal swing or a core update cannot be mistaken for your work.
When a class of fix keeps landing on your site and keeps producing nothing, it is retired for that site and stops appearing in your list. The recommendations get shorter, which is the point.
What it does for you
- Turns 847 findings into 3 decisionsGrouped by issue type and URL pattern, with the coverage that distinguishes a template from a handful of stragglers.
- Ranks them by what they are worthImpressions behind the affected pages, not a severity label invented in the abstract. The top row is the one to spend a Saturday on.
- Notices when you shipThe same watcher that spots a fix landing spots a template quietly breaking, which is a good reason to trust it.
- Proves it against pages you did not touchA frozen control set, measured over the same 28-day window. Both sides moving together is not a win.
- Says what to write next, from real demandA query with volume whose landing page never targets it is a page you have not written. Taken from what Google already shows you for, not from a keyword tool's guess.
- Finds the pages you are hiding by accidentEarning impressions with one internal link, or ranking despite a noindex nobody meant to leave in. Both are a fix measured in minutes.
- Stops suggesting what never worksRules that keep landing and keep producing no effect are retired for that site.
Nine lists from your Search Console
- Ready when you connect itQueries stuck on page two, positions clicking below what they should earn, pages losing traffic against the previous month, and two of your own URLs competing for one query. No crawl required, so they are the first thing you see.
- Ready after the first crawlDemand landing on a page that does not target it, pages earning impressions that nothing links to, short pages people are asking for, URLs ranking despite being excluded, and every under-clicked query with the reason named — a missing description, a duplicate title, a title of nine characters.
- Grouped by template, not by URLTwo hundred product pages with the same missing description is one row and one fix, the same way causes work. Every list exports to CSV.
One limit, stated plainly: the crawl stores a hash and a word count, never your page text. So these signals can tell you a page does not *target* a query — in its title, headings or URL — and never that it fails to *discuss* it. Telling you there is no page about something you have written three times is the kind of wrong that would cost us your trust.
Why the failures are the feature
There is no setting to hide a no-effect verdict, and adding one is not on the roadmap. Take the failures out and you have removed the reason to believe the wins — including from yourself, which is the reader who matters here.
Questions
What if there is no adequate control set?
The fix is recorded as landed with no clean comparison, and no verdict is issued. Small sites and sitewide template changes both hit this. Saying we cannot prove it beats a number you could not defend.
Does it replace my audit tool?
No, and we would rather say so. Research platforms find and size the work; this ranks it, watches for the deploy and measures the outcome. The overlap is the crawl, which is the cheap part of both.
How does it know I shipped something?
It compares a hash of each watched page's SEO-relevant fields against the previous check. The same machinery catches a template regressing, which is a good reason to trust it: it is not code that only looks for good news.
Try it on your own site, free
No card, and the free tier does not expire — the first verdict takes about six weeks to arrive.