Prooflytics
Strategy7 min read

AI Search Sends Clicks to Local Domains, Not Global Brands: What 87 Million Visits Show

An analysis of 87 million AI search visits across 10 markets found that clicks from AI-generated answers go to local domain variants, not global brand defaults. For performance marketers, this changes the geo-targeting and domain strategy calculus.

Global map visualization showing AI search traffic distribution across local domains in different markets

AI-powered search results send clicks to local domain variants, not global brand defaults, according to a Similarweb analysis of 87.6 million AI search visits across 10 markets by international SEO consultant Aleyda Solis. In markets from Germany to Brazil, local competitors outperform global brands in AI-generated click traffic even when the global brand has a dominant paid presence. For performance marketers running international campaigns, this changes the domain strategy and geo-targeting calculus.

Key takeaways

  1. Analysis of 87.6 million AI search visits across 10 markets found clicks go to local domain variants, not global defaults - Bol.com leads Dutch ecommerce over Amazon, MercadoLivre leads in Brazil.
  2. The number of domains capturing the first 50% of AI clicks varies enormously by category: a median of 5 in ecommerce, 17 in finance, and 47 in travel.
  3. Localization signals - local domain, local content, local entity mentions - appear to be ranking factors in AI search interfaces, not just traditional SERP.
  4. Campaigns targeting non-local domains in high-AI-search markets can expect lower click-through rates as AI interfaces preference local citations.
  5. B2B marketers with international audiences should audit whether their primary domain or a local variant receives AI search citations in their key markets.

What the 87-million-visit dataset shows

Localization signal definition: In AI search contexts, a localization signal is any factor that increases the likelihood an AI engine will cite a local domain variant (for example, bahn.de over booking.com, or mercadolivre.com.br over amazon.com) when responding to a market-specific query.

Aleyda Solis analyzed approximately 87.6 million AI search visits captured in March 2026 Similarweb data across the USA, UK, Spain, France, Germany, Italy, the Netherlands, Australia, Mexico, and Brazil - covering 57,696 domain-market combinations. The dataset captures visits that landed on a domain after a user clicked a link from an AI-generated answer in ChatGPT, Perplexity, Google AI Mode, or similar interfaces.

The finding that challenges the global-brand assumption: in ecommerce, the top-performing domain in most non-US markets was a local or regional player, not Amazon. In travel, local national rail and hotel aggregators outperformed global platforms like Booking.com in several European markets. Finance showed an even stronger local bias - local banks, local comparison sites, and government financial services portals outperformed global fintechs in citation frequency.

The market concentration data is particularly instructive. In ecommerce, 50% of AI search clicks in a given market go to just 5 domains. In finance, 17 domains share the first 50% of clicks. In travel, it takes 47 domains to capture 50% of AI search clicks. This means the category determines how concentrated the AI click opportunity is - ecommerce is a winner-take-most market for local AI citations, while travel is fragmented.

Why AI search interfaces preference local domains

The ICP problem this creates for global brands: paid media investments build Google Ads and Meta presence based on keyword and audience signals calibrated for traditional SERP behavior. AI search interfaces appear to weight different signals when selecting which sources to cite.

The most likely localization ranking factors in AI search, based on the Similarweb data pattern:

Local TLD and domain structure. bahn.de gets cited more than booking.com for German rail queries. mercadolivre.com.br gets cited more than amazon.com for Brazilian ecommerce queries. A local domain variant signals geographic specificity that AI models appear to use as a relevance proxy.

Local entity mentions. AI search models are trained on text that references local entities - local landmarks, local business names, local regulatory context. Content that naturally includes local entity references will be more closely associated with market-specific queries than content that is generically global.

Local language and content. Even when the global domain has a localized subdirectory (brand.com/de), the Solis analysis found that local domains (brand.de) and fully localized local-language domains outperformed subdirectory-based localization for AI search citations in most markets.

Prooflytics

Make the call with the whole picture

Briefs are daily; the understanding compounds.

14 days free · no credit card

Operational implications for performance marketers

This data changes three things for marketing teams running international campaigns.

First: Paid investment does not transfer to AI search clicks. A market where your global brand has 30% paid search impression share may have 3% of AI search citations. Paid and organic AI citation visibility are currently separate channels with separate ranking factors. Budget allocation for international markets needs to account for this decoupling - which also changes how you should measure AI search's impact on your marketing KPIs.

Second: Domain strategy affects AI visibility. If your company serves European markets from a .com domain with language subdirectories but lacks local TLDs, your AI search citation rate in those markets is likely lower than a local competitor with a market-specific domain. This is a structural disadvantage that cannot be overcome with content quality alone.

Third: Category concentration determines urgency. If you operate in ecommerce, the top-5 local domains capture 50% of AI search clicks. If you are not among those 5 in your key markets, the opportunity cost is high and growing. If you operate in travel, the fragmentation is higher and the urgency lower - but local positioning still matters.

Prooflytics surfaces channel-level traffic attribution in the daily briefing, enabling marketers to monitor when AI search referral traffic begins shifting between domains or increasing in volume relative to traditional organic. This makes the AI search visibility trend visible alongside paid channel performance without requiring a separate analytics setup.

What to audit in your domain and content strategy

Step 1: Check which domain receives AI search citations in your key markets. Search for your primary commercial keywords in ChatGPT, Perplexity, and Google AI Mode for each target market. Note which domain is cited - your global .com, a local TLD variant, or a local competitor. This is the baseline for your AI search positioning in each market.

Step 2: Audit your local TLD coverage. If your company operates in Germany, France, the Netherlands, Australia, Brazil, or Mexico but serves those markets from a .com domain, review whether acquiring and building out local TLD variants would improve AI search citation rates. The Solis data suggests this is a meaningful factor, particularly in ecommerce and finance categories.

Step 3: Add local entity signals to market-specific content. Content that references local regulatory bodies, local market statistics, local industry associations, and local geographic references is more likely to be cited by AI search for market-specific queries. Audit your market-specific landing pages and blog content for local entity density.

Step 4: Track AI search referral traffic separately. GA4 and most analytics platforms now surface AI search as a distinct referral source. Set up a segment or custom channel grouping for AI search referral traffic so you can see which domains and pages are receiving AI-generated clicks in each market. The broader picture of what this shift means for organic traffic is covered in the zero-click era analysis. This is the leading indicator of whether your localization efforts are moving AI search visibility.

Bottom line

  • 87.6 million AI search visits across 10 markets confirm that AI interfaces prefer local domain variants over global brands in most non-US markets.
  • Category determines concentration urgency: ecommerce (5 domains capture 50% of clicks) is highest urgency; travel (47 domains) has more fragmentation.
  • Audit which domain receives AI search citations in your key markets before assuming global brand presence transfers to AI search visibility.
  • Local TLDs, local entity signals in content, and local-language pages are the most actionable localization levers for AI search.
  • You can compare how marketing analytics platforms track AI search referral traffic and channel attribution on G2.

Frequently asked questions

Does this mean AI search ignores global brands entirely?+

No. The US market data in the Solis dataset shows that global brands with strong US-native presence (Amazon, Google, major US banks) do capture AI search citations at high rates. The local domain preference is most pronounced in non-US markets where a local competitor has a domain, content, and entity-signal advantage. In the US, the global brand often is the local brand.

Is the local domain preference specific to ChatGPT, Perplexity, or Google AI Mode?+

The Solis dataset aggregates across AI search interfaces, so it does not isolate the effect by platform. Google AI Mode may exhibit different localization weighting than ChatGPT or Perplexity based on how each platform handles geographic context in its search queries. The aggregate signal is clear - but platform-specific analysis requires your own testing.

How quickly can a local TLD domain build AI search citation volume?+

There is no established timeline, but the same factors that drive traditional SEO domain authority apply - content volume, inbound link signals, entity mentions from local publishers, and crawl accessibility. A new local TLD domain typically takes 6-12 months to accumulate the signals needed to compete with established local domains in AI search citation frequency.

Should B2B SaaS companies with international markets prioritize local TLD acquisition?+

For B2B SaaS, the travel and finance category concentration data is more relevant than ecommerce. In B2B, 47+ domains share the first 50% of AI search citations in most categories - the market is fragmented enough that strong content and entity signals on a .com domain with subdirectories can be competitive. Evaluate category concentration in your specific vertical before investing in local TLD acquisition.

How does this affect geo-targeting strategy in Google Ads?+

AI search and paid search are currently separate channels with separate click streams. AI search local domain preference does not directly affect your Google Ads auction performance or Quality Score. However, if AI search is diverting commercial-intent traffic to local competitors in your key markets, the total addressable click volume for paid search in those markets may be lower than historical benchmarks suggest.

Prooflytics

Make the call with the whole picture

Briefs are daily; the understanding compounds.

14 days free · no credit card

Continue reading

Strategy· 9 min read

Google Says Users Prefer AI Search. The Survey Evidence Is Less Clear.

Google regularly cites a Google/Ipsos survey as evidence that users prefer AI-powered search. SEO analysts have challenged the methodology, pointing to data gaps that inflate AI search preference claims. Before making channel strategy decisions based on this data, performance marketers should understand what the survey actually measures.

Strategy· 9 min read

How to Measure AI Search Impact on Marketing KPIs

AI search disrupts click-based traffic measurement but does not make marketing performance unmeasurable. Three approaches work in 2026: AI Visibility Tracking (citations and brand mentions in AI outputs), KPI replacement for click-based metrics, and revenue connection via incrementality testing. Here is how to implement each.

Strategy· 9 min read

Website Traffic Is Down 46%: What the Zero-Click Era Means for Marketers

Web traffic to brand and publisher websites has declined 46% over three years. AI search results, social feeds, and answer engines now satisfy user intent without a click. Performance marketers who keep optimizing for traffic volume are measuring the wrong thing. The shift is structural, and the channel mix that worked in 2022 no longer works today.

Strategy· 8 min read

LinkedIn Content and AI Search Citations: What 325,000 Prompts Show

LinkedIn is the second-most cited domain in AI search results, just behind Reddit. Semrush studied 325,000 prompts and 89,000 LinkedIn URLs. Company pages drive 59% of citations in Perplexity. Individual creators drive 59% of citations in ChatGPT and Google AI Mode. Structure and posting frequency matter more than likes.