What’s Going On: Recent Changes in How Google Counts Rank / Position / Impressions

Several recent shifts are causing confusion, fluctuations, and what many SEOs are calling “dirty data” in terms of rank tracking, impressions, and average position:

1. Google Limits Depth of Visible Results / Scraping Blocked
   • Tools that used to fetch many pages of SERP (positions 1–100) via scrapers are now being blocked or limited. Google is enforcing JavaScript rendering requirements, or making scraping harder.
   • As a result, many SEO tools now only reliably see the first page (top 10), or fewer results. Older rank data for deeper positions (e.g. 50-90) is no longer consistently captured.
2. Filtering of Impressions by Google Search Console (GSC)
   • Google appears to have adjusted when an “impression” is counted. Only if the search result is loaded and visible (or potentially loaded) on the page a user (or bot) is viewing. If your result is very deep (position 80+), or not in initial page load, it may not count. This leads to fewer overall impressions.
   • Because many impressions that used to come from extreme positions (which rarely lead to clicks) are dropped, the average position (as reported in GSC) tends to improve. The reason: the denominator (number of impressions) loses many poor-position impressions, so average shifts upward. This is the same idea described in your Spanish scenario.
3. Decoupling of Impressions and Clicks (“The Great Decoupling”)
   • As impressions go down (or change in quality), clicks do not always follow the same trend. In some cases impressions are rising in GSC for certain queries, but clicks are falling. Or vice versa. This makes CTR, average position, impressions less uniformly meaningful.
   • This is more pronounced in non-brand, informational queries: many impressions that never lead to clicks, or are simply not visible enough to be clicked.
4. SERP Feature Changes & AI/“Instant Answers”
   • More “zero result boxes”, “AI overviews”, featured snippets, knowledge panels, etc. These steal clicks and visibility from standard organic listings, changing the way users interact with results.
   • Also, Google’s shift in indexing, “crawled but not indexed” states, and quality filtering means some pages may no longer appear or be counted, even if they are technically present for bots.
5. Tooling & Reporting Lag / Variability
   • SEO tools that rely on scraping or SERP snapshots are having more outages, less depth, or inconsistent data due to Google’s anti-scraping efforts.
   • Search Console data itself is changing: Google is refining how and when “average position” is computed, which queries are considered, how “impressions” are defined.

Why Sites with Strong Non-Brand SEO Are More Affected

If much of your SEO performance is non-brand (i.e. targeting generic keywords, informational long tails, broad discoverability, rather than brand name searches), you’re especially exposed to these changes. Here’s why:

1. Many non-brand keywords rank very deep
   • Non-brand keywords tend to have less search volume, more competition, and often your pages will rank in positions well beyond page 1 or 2 (especially for newer content or niche topics). These deep rankings used to contribute impressions (even if low visibility), helping the “impressions count” and skewing average position. With those deep impressions now often not counted, the apparent visibility drops.
2. Low clickthrough at those depths
   • Even when those results are seen, they get negligible clicks. With impressions removed or filtered out, the few higher-position impressions dominate metrics like “average position”. So you see improved average positions on paper, but not much traffic change (because the real opportunity is still in those non-brand long tails which are now less visible in reports).
3. Reduced data for keyword tracking
   • If tools can’t reliably track beyond top 10 or 20 results (because of scraping limits or because Google intentionally doesn’t show deeper results to tools), you lose visibility into how far outside page 1 you are. That means you lose insight into the tail performance, which hurts strategic planning (e.g. content expansion, long tail SEO).
4. Impact on “discovery” vs “conversion” metrics
   • Non-brand traffic is often about discovery and volume. Google’s changes shift metrics to favor what is already performing well (brand, high volume, top SERP features). The long tail, which is harder to maintain and improve, becomes less measurable and more volatile.
5. CTR distortion
   • Because many impressions from low-visibility positions are gone, your remaining impressions tend to be in places where CTR is higher. That can skew your understanding of what’s working, maybe making your meta titles/descriptions seem more effective than they are, hiding issues for queries where you are still deep in the SERPs.

What This Means for You & Actionable Takeaways

Here are things you (with “much non-brand SEO” work) should watch out for, and some actions:

• Don’t panic at drops in impressions or changes in average position. They may not reflect worse SEO, but different reporting rules / less deep impression counting.
• Focus more on traffic, clicks, real visibility in top-10 or top-20 positions, not average position across all impressions.
• Use tools + Search Console together: see which queries are still ranking but are unmeasurable by some tools; watch for “crawled but not indexed” URLs.
• Improve content so that more keywords move from deep (page 2+) into page 1 – because these are the ones more reliably counted and visible.
• Optimize meta titles / descriptions for non-brand queries to improve CTR for the ones that are visible.
• Track SERP features & featured snippets, AI overviews etc., because often if your content is eligible, these show up and consume visibility.
• Re-evaluate keyword tracking approaches: maybe reduce the depth you try to monitor (e.g. focus on top 50 or 20 instead of 100), invest in tools that can handle Google’s JS/rendered SERPs, or that are more consistent.