Why is Google questioning the concept of position?

The SERP has exploded. It now stacks heterogeneous blocks: AI Overviews sometimes take up 40% of the screen, featured snippets divert organic clicks, and image or video carousels push traditional results further down. Five years ago, the SERP aligned ten blue links and position 3 meant something. This is no longer the case.

Mueller acknowledges this: the position has become a too simplistic indicator to reflect what the user really sees. Consider a result stuck under two PAA blocks and a featured snippet. Technically in position 4, it ends up below the fold, invisible without scrolling. As for generative AI elements, Google tracks them as an aggregated block, never specifying which occurrence produced which impression.

What does this concretely change in Search Console?

For content cited in AI Overviews, the Gen-AI performance report does not go down to the line-by-line position. You know your site has been included, you know the volume of impressions generated, but the exact location and prominence remain unknown. All citations from the same AI block are treated as a unique ranking entity.

This deliberate opacity complicates comparative analysis. First cited source or fifth? Complete excerpt or truncated? Impossible to decide. Click data remains accessible, but the position-CTR correlation becomes blurred, since CTR now depends on the density and nature of blocks located above the organic result.

Does Google have a solution to propose?

No, and Mueller admits this directly. Making these position metrics "useful" for site owners? He states that there is no simple solution. He even seeks suggestions from the SEO community, a revealing admission of a lack of ready answers.

Such unusual transparency reflects a strategic blur within Google itself. Product teams continuously test new SERP layouts, making any position metric inherently unstable. The call for contributions can also be read differently: a graceful way of saying "we won’t change anything, but we have listened to you."

• Modern SERPs are no longer linear lists, making the numerical position lose its meaning.
• Search Console does not detail the positions of citations in AI Overviews, only a global block.
• Google itself acknowledges it does not have a satisfactory alternative metric to propose.
• Visibility metrics (impressions, clicks) remain reliable, but their interpretation is becoming more complex.
• The average CTR by position loses its predictive value, as the SERP varies too much from query to query.

Is this statement consistent with ground observations?

Absolutely. For the past two years, SEO practitioners have observed that traditional positions no longer correlate with actual traffic as they used to. A site can gain three average positions on a cluster of keywords and lose traffic, simply because an AI Overview appeared in the meantime and captures 15% of clicks.

What Mueller does not say: this fragmentation of SERPs is a strategic decision by Google, which seeks to monetize more queries and keep users within its ecosystem. Presenting it as a neutral technical evolution is a convenient shortcut. [To be verified]: Google has not published any data on the rate of organic traffic cannibalization by AI Overviews, which would make this debate much more transparent.

What nuances should be added to this discourse?

Mueller speaks of "positions" in a broad sense. However, there are still pure transactional queries where SERPs remain quite traditional: 10 organic results, a few ads at the top. On these queries (purchase of specific products, local services), position 1 still performs significantly better than position 7.

Another nuance: Google has not given up on position as an internal ranking metric. Ranking algorithms still produce a linear order, and it is the user interface that fragments the display. Your content therefore retains an algorithmic position, simply masked or diluted by the presentation. Do not confuse ranking position with display position.

What risks does this opacity pose for SEOs?

The main risk? Losing the ability to diagnose finely a drop in traffic. If Search Console tells you "you have an average position of 8," but that this position 8 is sometimes under an AI Overview, sometimes under three PAAs, and sometimes without any block, you cannot isolate the real cause of the decline. The problem is methodological.

This leaves aggregated metrics (total traffic, conversions), which mask weak signals. You lose granularity in analysis, which favors large players who can afford alternative tracking (eye tracking, large-scale SERP scraping). [To be verified]: Google could provide a "block density" score per query to contextualize position, but no roadmap has been announced.

What should you concretely do with this information?

Stop focusing on average position in your client dashboards. Real impressions, CTR, and absolute organic traffic matter much more. A client asks you why you moved from position 4 to 6? Show them that traffic has increased anyway, with the disappearance of a competing featured snippet having freed up clicks.

Invest in tools that scrape actual SERPs, such as SEMrush, Ahrefs, or proprietary scripts, to map the blocks present on your key queries. Then build a custom indicator of "net visibility," which weighs the position by the presence or absence of competing blocks. More meaningful than a raw position from Search Console.

What mistakes should be avoided in light of this evolution?

The first mistake: neglecting AI Overviews on the grounds that Search Console provides no granular data. Manually monitor if your content is cited there, even without a precise metric. A citation in an AI Overview often generates fewer direct clicks, but it strengthens perceived authority and can trigger indirect backlinks.

Another trap: concluding too quickly that a rise in position without a rise in traffic is a bug or anomaly. A new block may have simply appeared above you. Always cross-check your position data with dated SERP screenshots.

How should you adapt your reporting and KPIs?

Offer your clients or management hybrid KPIs. “Organic traffic + Impressions in AI Overviews” as the main metric, instead of “Average Position.” Add a share of voice calculated on relative impressions against the competition: it handles SERP variations much better.

Also document SERP layout changes in your monthly reports, systematically: “In March, an AI Overview appeared on 40% of our target queries, explaining the drop in CTR despite stable positions.” Your discourse gains credibility and unjustified panics tend to fade away. If these analyses seem time-consuming or cumbersome to implement, seeking a specialized SEO agency can provide expert insight and advanced analysis tools to manage these new metrics accurately.

• Replace average position with organic traffic and impressions as the main KPI
• Regularly scrape target SERPs to map the blocks present
• Create a custom visibility indicator that integrates block density
• Manually monitor citations in AI Overviews, even without granular metrics
• Document SERP layout changes in monthly reports to contextualize variations
• Always cross-check Search Console data with dated SERP screenshots