Digital visibility in Switzerland in the age of AI Part 2: Zero-click, falling numbers and the new role of the website

In Part 1, we looked at the basics of AEO and GEO. Now we’ll focus on new phenomena such as the zero-click phenomenon, which is certainly shaking up your Google Analytics or Matomo statistics, and the new role of a company or organization’s website.

Digital visibility is undergoing a profound upheaval that is neither temporary nor reversible. The question is not whether the rules of the game will change – they already have. The question is how quickly and consistently companies will react. This development certainly offers opportunities for Swiss SMEs. The strengths that characterize many SMEs – deep expertise, regional roots, personal customer relationships, quality orientation – are precisely the characteristics that make generative AI systems trustworthy. Those who make these strengths digitally visible – through substantial content, a professional web presence and consistent E-E-A-T signals – will also be found, quoted and contacted in the AI era.

If you’ve taken a look at your Google Analytics or Search Console data in recent months, you may have noticed a worrying trend: Organic clicks are dropping even though your rankings have remained stable or even improved. This phenomenon is not a sign that your SEO work has failed. It is an expression of a structural change that is affecting the entire industry.

The term “zero-click search” describes search queries where users receive their answer directly on the search results page – through featured snippets, knowledge panels, Google AI overviews or other SERP elements – and never click on an organic result. Current surveys by various analysis providers indicate that around 60 percent of all Google searches already end without a click. For search queries that trigger an AI Overview, this figure is even significantly higher, in some cases at over 80 percent. Forecasts for 2026 assume a further increase.

For Swiss SMEs, this means that even a position 1 ranking below an AI overview delivers significantly less traffic today than it did two years ago. The search results page has evolved from a directory of external links to an information platform in its own right. Google answers the question – visiting your website becomes optional.

Before you panic: falling visitor numbers are not automatically synonymous with a decline in business. There is a crucial difference between traffic volume and traffic quality. Users who click on your website despite the presence of AI responses do so with a more specific intention – be it a purchase decision, making contact or a deeper need for information. Initial analyses by international providers show that visitors who access websites via AI platforms such as ChatGPT or Perplexity are significantly more likely to convert than traditional organic traffic.

This does not mean that traffic has become irrelevant. It means that the previous equation “more visitors = more business” no longer applies without restriction. If you only look at visitor numbers, you are missing the real picture.

Not all content is losing clicks to the same extent. Informational content that can be summarized in a few sentences is most affected: Definitions, simple instructions, factual queries, FAQ answers. In other words, precisely the content that AI systems are best able to synthesize. Less affected are transactional search queries (users want to buy or book), navigational searches (users are looking for a specific website) and complex decision-making searches that require personal advice or comparisons.

This has a clear implication for Swiss companies: anyone who has primarily filled their website with informational content in order to gain rankings must rethink their content strategy. Not in the direction of “less content”, but in the direction of “other content” – content that AI systems cannot replace, but can only reference.

In view of the zero-click development, an obvious question arises: Do we still need our own website at all? The answer is a resounding yes – but the rationale has shifted.

In the past, the website was primarily a traffic channel: Visitors came via Google, gathered information and became customers. Today, the website is also – and increasingly primarily – a signal of trust. AI systems such as Google Gemini, ChatGPT or Perplexity search the web for sources that they consider trustworthy in order to generate their answers. The quality, structure and authority of your website determine whether your content is used as a source and cited.

A well-structured website with clear author information, comprehensible source references and up-to-date, in-depth content is therefore not just a sales tool, but a prerequisite for being present in the AI-driven information landscape at all. This applies to both the texts themselves and the technical basis: clean HTML, structured data (Schema.org), fast loading times and a barrier-free, mobile-optimized presentation.

Anyone planning a website relaunch in 2026 or commissioning a new website should consider this changed role from the outset. It is no longer enough to create an aesthetically pleasing site with general company texts. A modern relaunch must also take into account how AI systems read and evaluate the content.

In concrete terms, this means that each page should have a clearly defined function and be geared towards a specific user intention (search intent). Structured data must be implemented from the outset – not as an SEO add-on at a later date, but as an integral part of the information architecture. Author information, company details and contact data should be stored in a machine-readable format. And the content itself must be formulated in such a way that it offers maximum benefit for both human readers and AI crawlers.

E-E-A-T in practice: building trust that machines understand

The acronym E-E-A-T stands for Experience, Expertise, Authoritativeness and Trustworthiness – in other words, experience, expertise, authority and trustworthiness. Google has been using this framework in its Quality Rater Guidelines for years, but with the advent of generative AI, E-E-A-T has taken on a new urgency. This is because AI models give weight to precisely those signals that E-E-A-T describes when selecting their sources.

For Swiss SMEs, which have neither large editorial teams nor international brand awareness, this may seem daunting at first. However, E-E-A-T actually offers smaller companies in particular an opportunity – because experience and expertise cannot be bought, they have to be acquired and documented.

Experience
Show that there is real practical experience behind your content. This can be done through documented case studies and project reports, before-and-after illustrations, concrete figures and results from customer projects or testimonials in your own words (not generic testimonials). A Swiss trade company that describes in detail how it solved a specific renovation problem generates more E-E-A-T signals than a generic performance page with stock photos.

Expertise (professional competence)
Professional competence is reflected in the depth and accuracy of the content. This does not mean that every text has to be a scientific paper – but it must show that the author knows what he or she is talking about. Specifically: Use industry-specific terminology where it is appropriate. Explain complex issues precisely but clearly. Cite sources when quoting data or statistics. And – particularly relevant – identify your authors by name and link to their qualifications.

Authoritativeness (authority)
Authority is not created on your own website alone, but through the interplay of various signals: mentions in industry media, links to thematically relevant pages, articles in specialist publications, memberships in associations or certifications. Regional authority signals are particularly effective for Swiss SMEs – such as mentions in local media, cooperation with universities of applied sciences or partnerships with industry associations.

Trustworthiness
Trustworthiness is the overarching signal into which the other three factors feed. Technically, it is demonstrated by HTTPS encryption, a correct legal notice (required by law in Switzerland), a transparent privacy policy and clearly recognizable contact options. In terms of content, it is demonstrated by an honest approach to one’s own offer – i.e. also by naming limits, by fair price communication and by refraining from exaggerated advertising promises.

Overview: E-E-A-T for Swiss SMEs

The content strategy of many websites is still based on a model that was developed for classic search engine optimization: create as many pages as possible for as many keywords as possible in order to increase the probability of organic rankings. This model is losing its effectiveness – not because keywords have become irrelevant, but because the visibility of individual pages increasingly depends on factors that go beyond pure keyword coverage.

Probably the most important shift in the 2026 content strategy can be summarized in one sentence: Fewer pages, but better ones. Many companies – including Swiss SMEs – have produced large amounts of content in recent years, often with the help of AI text generators that deliver grammatically correct but interchangeable content. This content may have generated short-term rankings, but generative AI systems recognize generic content and prefer sources that offer real added value.

The recommendation is therefore: Reduce the number of your pages if necessary. Consolidate thematically similar content into comprehensive, in-depth resources. Make sure that each page has a clear reason why a user should read it – and that this reason is recognizable in the first few lines.

There are certain content formats that cannot be structurally replaced by AI responses. Those who align their content strategy with this will ensure long-term relevance. These include: original data and studies (own surveys, market observations, industry figures), interactive tools and calculators (cost calculators, configurators, calculation tools), personal perspectives and opinion pieces with comprehensible arguments, in-depth analyses with a regional focus and multimedia content such as videos, podcasts or documented workshops.

A Swiss trustee who publishes an annual analysis of the cantonal tax burden for SMEs produces content that no AI system can generate out of thin air – because the data is original. It is precisely this type of content that generative systems use as sources and cite.

The use of AI tools for content creation is not problematic per se. It becomes problematic when AI-generated content is published without human refinement. Machine-generated texts serve as source material that must be enriched by experts with their own knowledge, concrete examples, regional references and a recognizable attitude. The formula is: AI as an efficiency tool, humans as a guarantee of quality.

Google Business Profiles and regional signals

While the global search landscape is being revolutionized by zero-click and AI answers, one area remains comparatively stable: local search. For Swiss SMEs whose business is tied to one location – whether it’s a craft business in Burgdorf, a dental practice in Biel or a boutique hotel in Interlaken – local visibility is still a key lever for growth.

The Google Business Profile (formerly Google My Business) is the most important tool here. A fully completed, regularly updated profile with photos, reviews, opening hours and categories not only influences the ranking in the Local Pack (the map results), but also provides data that AI systems use to generate local recommendations.

In addition to the Google Business Profile, there are other measures that strengthen local visibility: consistent NAP data (name, address, telephone number) across all online directories – from local.ch to search.ch and industry-specific platforms. Local backlinks from community websites, regional media or local partner organizations. Content with an explicit regional reference that provides clear location signals not only for Google, but also for AI systems. And reviews that are actively solicited and professionally answered – because reviews are one of the strongest trust signals in a local context.

The Swiss market structure brings with it specific characteristics that must be taken into account in the local search strategy. Multilingualism (German, French, Italian, Romansh) means that a company in Biel/Bienne should ideally maintain its Google Business Profile in two languages. Cantonal differences in regulations, designations and search habits require locally adapted content. And the high purchasing power and quality orientation of Swiss consumers in an international comparison makes trust signals such as ratings, certificates and association memberships particularly effective.

What to measure when clicks are no longer enough?

In Part 1, we mentioned measurability as a challenge. Here, we take a closer look at which key figures will actually be meaningful in 2026 and how modern monitoring can be set up.

From traffic KPIs to visibility KPIs
The classic SEO metrics – organic sessions, click-through rate, page views – remain relevant, but only tell part of the story. In addition, new metrics need to be established that measure visibility even where no click takes place.

A particularly revealing new KPI is Branded Search Volume – the number of search queries that contain your brand name. Why? When users Google your company name, they have already noticed your brand – possibly through an AI mention, a social media presence or a personal recommendation. Branded searches are therefore an indicator of the effectiveness of your overall visibility strategy, regardless of whether the first contact was via Google, ChatGPT or a LinkedIn post.

Monitoring whether and how your brand or your content appears in generative AI responses is not yet technically trivial. There are initial tools and approaches – such as manual tests with ChatGPT, Perplexity or Google Gemini for relevant industry queries, specialized monitoring platforms that track AI citations, and regular evaluations of the Search Console for changes in impressions and CTR. This field is developing rapidly and it can be assumed that significantly more sophisticated monitoring solutions will be available in 2026 and 2027. Until then, a pragmatic approach is recommended: regularly test relevant search queries in various AI systems and document whether and how your website appears as a source.

Recommendations for you as a Swiss SME

The following recommendations summarize the key findings from Part 1 and Part 2 of this series and are specifically tailored to the needs of small and medium-sized enterprises in Switzerland.

Take falling visitor numbers seriously, but don’t get carried away. Check whether the remaining traffic is of higher quality. Measure conversions, not just sessions.

Invest in your website as an anchor of trust. Make sure that your site is structured in a technically flawless, substantial and machine-readable way. Schema.org markup, clear author information and up-to-date content are not optional, but mandatory.

Build up E-E-A-T in a targeted manner. Document your experience, demonstrate your expertise, strengthen your authority through networking and industry presence, and communicate in a trustworthy manner.

Rethink your content strategy. Less generic content, more original data, regional analysis and personal perspectives. Create content that AI systems need as a source, not content that can replace them.

Maintain your local visibility. Google Business Profiles, consistent NAP data, local backlinks and active review management remain highly effective.

Establish new KPIs. Supplement your reporting with branded search volume, SERP impressions, AI mentions and engagement metrics.

Consider visibility as an overall strategy. SEO, AEO and GEO are not isolated disciplines, but facets of an integrated digital visibility strategy. Supplement this with a social media presence, newsletter marketing and – where appropriate – paid advertising.

Bibliography

Bain & Company, Goodbye Clicks, Hello AI: Zero-Click Search Redefines Marketing, Bain Insights, 2025.

Similarweb / Semrush, Zero-Click Search Data 2024-2025, various analyses.

Onely.com, Zero-Click Search Is Evolving Into Zero-Search Discovery, December 2025.

Digiday, AI is driving more traffic, but not offsetting zero-click search, July 2025.

Ekamoira Research, 60% of Searches Get Zero Clicks: How to Win in 2026, January 2026.

Google, Search Quality Evaluator Guidelines (E-E-A-T), updated 2025.

Chen et al, Generative Engine Optimization: How to Dominate AI Search, arXiv 2509.08919, 2025.

Digitalmarketing.gmbh, Trends 2026: Why SEO is not dying but becoming a data source for AI, Dec. 2025.