Breaking: Rank tracking is cracking. After weeks of quiet shifts in Google Search, the old ways to measure position and visibility are failing. I ran fresh tests across multiple markets this morning. The data confirms a pattern. Rankings look stable one hour, then scatter the next. Impressions spike, but clicks slide. The tracker you trust is now the weakest link.
What changed, and why it matters
In mid September, Google removed the option to show 100 results per page. That single UI change blew a hole in many rank tracking systems. Tools that used to grab deep result sets now need many more queries to reach the same depth. That means higher costs, more throttling, and more noise. At the same time, AI Overviews and new result modules now sit above or within the classic list. They change what users see, and where they click.
In my controlled checks, AI led blocks often push the first organic blue link far below the fold. That means more zero click answers. It also means your average position can drop even as impressions rise, since more surfaces are counted and more result types appear. Search Console is adding smarter filters and clusters, which helps analysis. It does not fix the basic problem, the search page itself is now fluid, personalized, and stitched from many parts.
[IMAGE_1]
Historic comparisons are broken. Do not treat pre September rank and CTR baselines as apples to apples with today.
Why traditional tracking is failing
Classic trackers assume a fixed stack of links and a stable page depth. That world is gone. There are three main breaks.
First, the missing 100 results per page option forces shallow sampling. You need more queries to collect the same slice of reality. Many tools are not doing that yet.
Second, the page is no longer a page. AI blocks, carousels, and mixed packs vary by user and moment. Position 3 can be 1,000 pixels down one hour, then 400 pixels later. Your number is not your placement.
Third, reporting logic is in flux. I am seeing more impressions counted with lower average positions. That points to changes in how results are counted and displayed. It also points to AI surfaces taking attention away from classic links.
A playbook for SEO teams, starting now
Here is what to do, in order.
-
Stabilize your tracking stack
Align query depth per market and device. Increase the number of locations tested. Parse modules, not just links. Use official APIs where possible. Expect higher query volume and higher bills. Document new limits and sampling rules. -
Shift your KPIs
Move from rank to outcomes. Focus on clicks, CTR by page, and conversions. Track position in buckets, for example top 3, 4 to 10, 11 to 20. Add a visibility score that combines pixel depth and result type. Watch branded and non branded segments apart. -
Level up Search Console use
Set country and device filters on every view. Use query groups to separate intent classes. Annotate the dates of major search changes. Compare pages against their own past, not site wide averages.
[IMAGE_2]
-
Format for AI Overviews
Lead with a crisp answer in the first 2 sentences. Use clear headings and short sections. Add FAQ blocks and how to steps. Mark up products, authors, and organization with structured data. Include concise images that explain the answer. -
Rebuild reporting baselines
Start a new baseline from October 2025. Flag a confidence range on all rank and CTR charts. Explain the new rules to leadership. Tie targets to click share and revenue, not a single position number.
Track what you can control. Content quality, answer clarity, and technical health now beat chasing a single rank line.
What this means for teams and users
Budgets will rise for data collection. Some tools will lag or break. Expect weekly reports to carry more caveats, and more focus on outcomes. Users will get faster answers at the top, often without a click. Publishers that structure answers well will keep visibility. Those that rely on position alone will lose ground.
- Tool costs go up, depth goes down, unless you change methods.
- Executive dashboards must pivot to click share, not position.
- Content teams need AI friendly formats and tighter intros.
- Year over year charts need new baselines and clear notes.
Frequently Asked Questions
Q: Why did my impressions rise while average position fell?
A: More result types are counted, and AI blocks push links down. You are seen more, but clicked less.
Q: Can I still trust rank trackers?
A: Only if they updated their methods. Check their query depth, location mix, and module parsing. Ask for documentation.
Q: What should I report to executives now?
A: Report clicks, CTR by page, conversions, and visibility by intent. Include position buckets with clear caveats.
Q: How do I optimize for AI Overviews?
A: Lead with a direct answer, cover key entities, and use structured data. Add short FAQs and step by step sections.
Q: Is this permanent?
A: The direction is clear. AI led results will expand. Plan for this new normal.
The headline is simple. Tracking, as we knew it, is over. The winners will measure what users actually do, not just where a link sits. Build for answers, track for outcomes, and reset how you prove impact. The sooner you adapt, the less your data will lie to you.
