SISTRIX this month published its February 2026 monthly review, presenting data from an analysis of over 100 million keywords in Germany that quantifies for the first time the precise traffic cost of Google's AI Overviews on the German search market. The figures land roughly one year after AI Overviews became broadly active in Germany and document losses that vary dramatically by industry, with certain health and information publishers absorbing click reductions that dwarf the average.

The review, sent by Johannes Beus, founder and managing director of SISTRIX, covers three separate subjects: the AI Overviews click-through rate impact in Germany, Google's apparent algorithmic downranking of self-promotional listicles, and a SectorWatch analysis on the vinyl record market. The Bonn-based analytics company has been tracking the German search landscape for 18 years - its Sichtbarkeitsindex (Visibility Index) turned 18 last week.

The headline numbers

According to SISTRIX, AI Overviews are served on approximately 20% of all keywords in Germany. That reach translates into a direct and measurable cost: the analysis puts total lost organic clicks at 265 million per month across the German market.

The CTR decline at the top of the results page is steep. According to the analysis, the click-through rate for the first organic position falls from 27% to 11% when an AI Overview is present - a loss of nearly 60% of the clicks a page in position one would otherwise receive. Averaged across all keywords, including those where AI Overviews do not appear, the mean click loss sits at 6.6%.

The scale of this finding is significant in context. Earlier studies from other markets offered comparable signals. Research from Ahrefs published in April 2025 found a 34.5% reduction in organic clicks for top-ranking sites when AI Overviews are present, based on an analysis of 300,000 keywords. Seer Interactive published data in November 2025 showing organic CTR for informational queries fell 61% since mid-2024, across 3,119 search terms at 42 organizations. The SISTRIX figure of a 59% decline at position one for the German market is consistent with the upper end of that range.

Industry-level divergence

The 6.6% average figure conceals a wide distribution. According to SISTRIX, click losses by industry range from 1% at one extreme to over 24% at the other. The pattern is not random. Categories built on informational intent - health questions, parenting, home improvement - are the hardest hit. These are precisely the queries where AI Overviews provide a direct answer and eliminate the motivation to click further.

The single largest absolute loser identified by the analysis is Wikipedia, which according to SISTRIX loses 31.6 million clicks per month as a result of AI Overviews in Germany alone. Specialised health portals represent the biggest proportional losers, with some recording over 30% click declines.

That pattern has parallels across markets. A January 2026 report covering health publishers found authoritative medical information sites that had held dominant positions for years losing significant visibility following Google's December 2025 core update. The intersection of algorithmic changes and AI Overview expansion is creating compounding pressure on informational content publishers.

What this means for keyword analysis

The SISTRIX review makes a methodological argument that goes beyond the click numbers themselves. According to Beus, search volume as a standalone metric for evaluating keywords has become insufficient. Whether AI Overviews appear for a given keyword - and how aggressively they suppress clicks - must now be factored into any keyword prioritisation decision.

That reframing has operational implications. A keyword with 50,000 monthly searches where AI Overviews appear for 80% of queries may deliver far less traffic than a keyword with 20,000 searches where the feature rarely triggers. Tools that do not surface this distinction are providing an incomplete picture of opportunity. Previous research from SISTRIX released in August 2025 had documented AI Overviews appearing for 17% of German keywords - a figure the new analysis now places closer to 20%, indicating continued expansion of the feature's footprint.

The challenge for publishers sits between two strategies. Producing content that an AI summary cannot replace - original research, primary data, lived experience - maintains a reason for users to click through. The alternative path is optimising to be cited as a source within the AI Overview itself, which SISTRIX's own October 2025 analysis of German AI Mode found to be heavily skewed toward Google-owned properties, with YouTube capturing 40% of AI Mode citations and Google.com itself receiving 31%.

Self-promotional listicles under pressure

The February review also addresses a separate development: Google appears to be significantly reducing the visibility of websites that publish listicles in which they rank themselves first. The tactic is familiar to most SEO practitioners. A SaaS company publishes an article titled "The 10 Best CRM Tools in 2026," lists itself at number one, updates the year periodically, and benefits from the freshness signal that "best" queries typically reward. According to SISTRIX, SEO analyst Lily Ray examined several cases following ranking fluctuations in January 2026 and found that websites using this approach lost up to 49% of their visibility.

The affected websites share a set of characteristics beyond the listicles themselves. According to the SISTRIX analysis, they show heavily scaled content output, AI-generated text without editorial oversight, artificial freshness updates, and schema markup misuse. All identified cases come from the SaaS sector. The concern for publishers extends beyond traditional search. According to SISTRIX, losses in organic results carry over into Gemini, AI Mode, and ChatGPT - platforms that draw from the same quality signals when deciding which sources to surface.

PPC Land has previously documented this self-promotional pattern in AI Overviews, reporting in May 2025 how AI Overviews had become a target for manipulation, with listicles from company websites appearing in AI-generated answers despite representing self-interest rather than independent assessment. Ray had flagged similar dynamics at the time, raising concerns about whether undisclosed commercial relationships might be driving some citation patterns. The January 2026 ranking volatility appears to have precipitated a more direct correction.

The wider enforcement context matters. Google in late 2024 linked organic search penalties directly to advertising eligibility for the first time, a structural change that raised the stakes of content quality violations across its ecosystem. Separately, in December 2025, the company deployed its third core update of the year, which brought sustained volatility to many categories of content. Whether the January 2026 listicle visibility drops represent a discrete algorithmic signal targeting this specific behaviour or part of that broader continued volatility is not confirmed by Google. SISTRIX presents them as a coherent pattern consistent with a targeted response.

Lily Ray, analysed by PPC Land in December 2025, has observed that over-optimised content strategies are increasingly the cause of ranking losses rather than their cure - a position this latest evidence reinforces.

The vinyl sector as a search case study

The February review includes a SectorWatch analysis by Helen Schrader examining how the German vinyl record market maps onto search behaviour. The sector is large enough to be instructive: according to SISTRIX, 4.9 million LPs were sold in Germany, generating 162 million euros in revenue, as the format holds its own against streaming. Schrader analysed 382 transactional keywords and 408 informational keywords.

The findings describe a clean separation between the types of domains that win in each category. According to SISTRIX, the top domains in transactional search and informational search barely overlap. Winning both simultaneously through a generalised approach is not what the data shows. Specialist retailers dominate transactional results; editorial and community sites dominate informational queries. Longtail queries such as "Schallplatte knistert" (record crackling) or questions about pressing matrix numbers for specific albums carry little measurable search volume but represent the moments where audience trust is built. Content that explains why the same album costs 2,000 euros in one pressing and five euros in another qualifies buyers before they are ready to purchase. SISTRIX frames this as pre-qualification rather than traffic acquisition.

The vinyl analysis offers a sector-specific illustration of the same structural pressure documented in the AI Overviews data. Categories built on answering questions are increasingly competing with Google's own AI-generated responses. Transactional intent, by contrast, remains a category where organic results hold more ground, partly because product queries involve price, availability, and trust signals that AI summaries cannot fully resolve.

Google Discover and the Googlebot 2MB limit

The review closes with two additional technical notes. According to SISTRIX, in early February 2026 Google rolled out the first standalone Discover core update - a dedicated algorithmic adjustment specifically targeting the Discover feed rather than folding Discover changes into broader core updates. According to SISTRIX, the update penalises pure clickbait strategies and increases weighting for local relevance and genuine topical authority.

PPC Land covered the February 2026 Discover core update in detail on February 5, reporting that it introduced three changes: prioritising locally relevant domestic content, reducing sensational material, and elevating deep expertise from specialised sections of broader publications. The geographic implications prompted concern from international publishers, with industry executives describing the filtering as potential digital discrimination against non-US sources.

The review also flags a Googlebot crawling limit that affects any publisher operating large sites: Googlebot processes only the first two megabytes of uncompressed HTML per page. Content, internal links, and structured data that appear beyond that threshold are invisible to Google's crawler. For sites running extensive navigation, lengthy footers, or late-loading content blocks, this boundary is not theoretical.

Context for the marketing community

The 265 million monthly click figure for Germany is not a projection or model estimate - it is presented by SISTRIX as a measurement derived from over 100 million keywords and their actual CTR data with and without AI Overview presence. The methodology builds on the same infrastructure that produced the August 2025 finding of 17% AI Overview penetration in German search, a figure now revised upward to approximately 20%.

For marketing professionals managing organic search strategies in Germany or comparable European markets, the data confirms a shift that multiple studies have now documented across different geographies. The click loss is real, it is uneven, and it is most severe for publishers who built their traffic on informational content with straightforward answers. The category of content least affected, according to the SISTRIX pattern, is content that requires experience, primary data, or transactional context that AI summaries cannot replicate.

The SISTRIX Visibility Index, which the February review notes has just turned 18 years old, has expanded its scope from Google search rankings to include Amazon visibility and AI system presence. The company will present further findings on AI search at the OMT Summit in Düsseldorf on 5 and 6 March 2026, where Beus is scheduled to speak on 6 March at 11:45 on Mainstage 1 about what over 100 million prompt responses reveal about competition between Google and OpenAI.

Timeline

Summary

Who: SISTRIX, a German SEO analytics firm founded and led by Johannes Beus, published the February 2026 monthly review. Findings on self-promotional listicles draw on analysis by SEO analyst Lily Ray. The SectorWatch section was authored by Helen Schrader.

What: An analysis of over 100 million keywords in Germany found that AI Overviews reduce the click-through rate at position one from 27% to 11%, cost the German market 265 million organic clicks per month, and affect 20% of all keywords. Separately, sites publishing self-promotional listicles in the SaaS sector have lost up to 49% of their search visibility following January 2026 ranking changes.

When: The monthly review was sent on 1 March 2026, covering developments from February 2026 and drawing on approximately one year of AI Overview presence in Germany.

Where: The analysis focuses on the German search market specifically, though the patterns observed - CTR decline, health sector losses, SaaS listicle penalties - align with dynamics documented across other markets including the UK, US, and India.

Why: The data matters because keyword analysis built on search volume alone no longer predicts traffic reliably in markets where AI Overviews are active. Publishers and marketers operating in Germany or comparable European markets need to factor AI Overview prevalence into planning. The self-promotional listicle findings indicate that a once-common SEO tactic has become a liability, with losses extending beyond Google into AI-powered search platforms including Gemini and ChatGPT.

Share this article
The link has been copied!