Indexed pages keep falling
Search Console's indexed count slides week over week while nothing obvious changed. Google is quietly losing confidence in whole sections of the site.
When traffic collapses, guessing is expensive. I find the root cause, brief your leadership in plain English, and direct your developers through the fix until the comeback shows in the data.
SEO recovery is forensic work: tracing a traffic or indexation collapse back to its root cause, directing the fix, and proving the comeback against a baseline. A routine audit surveys everything a site could improve. A recovery engagement answers one question: what exactly broke, and when.
The distinction matters because a collapsing site does not need fifty recommendations. It needs the one that ends the bleeding. If nothing has collapsed and you want the broad survey, you want an SEO audit. If clicks or indexed pages fell off a cliff, you are in the right place.
Guessing is what makes traffic drops expensive. Teams rewrite content, buy links, and switch agencies while the actual cause sits in a redirect map nobody has opened in years. I have been finding these causes since 2004, and the pattern repeats: the symptom is recent, the cause usually is not.
Symptom check
If any of these look familiar in your Search Console, the problem is diagnosable. These are the five I see most.
Search Console's indexed count slides week over week while nothing obvious changed. Google is quietly losing confidence in whole sections of the site.
Clicks fell off a cliff on a date you can point to. That date is evidence: whatever shipped, migrated, or expired that week is suspect number one.
The new site launched, then rankings slid for weeks. Usually mechanical: lost redirects, changed URLs, rewritten titles.
How migrations should work →Google fetches your pages and declines to keep them. When that status grows, Google increasingly sees the site as not worth indexing.
Pages returning 200 that Google reads as empty. Common with app shells that render blank for crawlers, and every one erodes sitewide trust.
From a real diagnosis
Tap the numbered markers. This is the shape of a real recovery: a B2B SaaS whose indexation flatlined for reasons that started two years before the drop.
Five steps, always in this order. Diagnosis before prescription, evidence before invoices.
Date the break precisely. Search Console, analytics, deploy logs, and archived crawls narrow the collapse to a week, often to a day. The date is the single most valuable clue in the whole engagement.
Crawls plus redirect and canonical archaeology (the cause is often years older than the symptom). I follow the evidence backward until the explanation accounts for the whole curve, not just part of it.
A written diagnosis in plain English, addressed to the CEO rather than the dev team: what broke, when, why, what it is costing, and the fix in dependency order. No jargon, no fifty-item list.
Your engineers ship the fixes. I set the order, review each change, and verify it landed the way Google needs it to. Nobody rebuilds anything that does not need rebuilding.
Indexation and click curves tracked against the pre-drop baseline until the comeback is visible in the data. You get proof, not a promise that things feel better.
the year I started diagnosing ranking collapses. The failure patterns repeat; the platforms just rename them
how far back the root cause hid in the featured SaaS case before the symptom ever showed
root cause per collapse, almost every time. Everything else on the list is a symptom
Fixed-fee diagnosis, quoted before we start. Monitoring the comeback afterward is an optional retainer.
Most recovery clients are founders, CEOs, and marketing leads who have already heard three theories from three people and cannot tell which one to fund. What you get from me is a diagnosis in plain English you can take to a board, and a fix directed to completion instead of a report that dies in a drawer.
I do not replace your engineering team. I direct it: what to fix, in what order, and how to verify each change landed the way Google needs it to. Search forensics is a different discipline from building features, and it should not be learned on a live emergency.
This is the most recoverable category there is, because the causes are mechanical and they leave fingerprints. It is the same discipline my website migration SEO service applies before a launch, pointed backward at one that shipped unprotected.
The diagnosis is a fixed fee, quoted after a short scoping call. If you want the comeback watched through to completion, monitoring afterward is an optional retainer. No percentage of ad spend, no lock-in, and no invoice for guessing.
Straight answers about traffic drops and what it takes to come back.
Send me a screenshot of your Search Console performance curve and I will tell you on the first call whether the drop looks mechanical, algorithmic, or structural, and what a diagnosis would take. No retainer required to find out what broke.