Key finding: In a 28-day tracking study across 82 GEO and AEO companies, 66% received zero AI citations across four major AI engines (ChatGPT, Perplexity, Gemini, and Claude). This is the GEO Visibility Gap: the majority of the market — including many agencies that sell AI visibility services — cannot be found by the AI engines they claim to optimise for.
Background and Methodology
From 14 May to 11 June 2026, Kaus AI ran a structured daily tracking programme across 82 companies operating in the GEO, AEO, and AI-search visibility space. Companies included pure-play GEO agencies, full-service digital agencies with GEO offerings, and GEO/AEO software platforms.
How it worked:
Each day, Kaus AI's tracking pipeline ran four buyer-intent queries across four AI engines:
- "What are the best GEO (Generative Engine Optimisation) companies or tools?"
- "Best AEO (Answer Engine Optimization) agencies and tools?"
- "Which tools help brands appear in AI search results?"
- "Top AI visibility tracking platforms for brands?"
Engines queried: ChatGPT (GPT-4o), Perplexity AI, Gemini (Google), Claude (Anthropic)
Total data points collected: 13,000+ individual company–query–engine–date combinations
Classification: Each response was parsed for company mentions and coded as mentioned (1) or not mentioned (0), with position and context logged where applicable.
Key Findings
Finding 1: 66% of tracked companies have zero AI citations
Of 82 companies tracked over 28 days, 54 (66%) received no citations at all across any engine, any query, on any day. These companies run GEO or AEO services but are absent from the AI responses their target buyers see.
The remaining 28 companies (34%) received at least one mention across the tracking period.
What this means: The GEO market has a strong visibility concentration effect. Two-thirds of participants in the space are functionally invisible to the AI engines they operate in.
Finding 2: Citation is heavily concentrated in the top 5
The top 5 most-cited companies account for approximately 86% of all citation events in the tracking period:
| Rank | Company | Total Citation Events |
|------|---------|----------------------|
| 1 | Semrush AI Toolkit | 232 |
| 2 | Ahrefs Brand Radar | 154 |
| 3 | Profound | 38 |
| 4 | Writesonic GEO | 25 |
| 4 | Otterly.AI | 25 |
| 5 | Peec AI | 24 |
| 6 | First Page Sage | 22 |
| 7 | iPullRank | 18 |
| 8 | Omnius | 12 |
| 9 | Omniscient Digital | 9 |
| 10 | Intero Digital | 8 |
Companies with no citations: 54 of 82 (66%)
Finding 3: Tools are cited 4× more than service agencies
Software platforms and SaaS tools dominate AI citations. Semrush, Ahrefs, Profound, Otterly.AI, and Peec AI — all monitoring or optimisation platforms — account for the top 5 positions and the majority of all citation events.
By contrast, pure-play service agencies (Minuttia, iPullRank, First Page Sage) appear in lower positions with fewer citation events. Among 82 tracked companies, the tool-to-agency citation ratio is approximately 4:1.
What this means: If you are building a GEO service agency, your brand is competing in a category where software tools have a natural citation advantage. Building a software product alongside a services practice significantly increases AI citation frequency. Agencies without a tool layer face a structural disadvantage in AI engine visibility.
Finding 4: Query type significantly affects who gets cited
Not all query types produce the same citation set:
| Query | Total Mentions across tracking period |
|-------|--------------------------------------|
| "Top AI visibility tracking platforms for brands?" | 165 |
| "Which tools help brands appear in AI search results?" | 159 |
| "Best AEO agencies and tools?" | 153 |
| "Best GEO companies or tools?" | 130 |
The "AI visibility tracking platforms" query generated the most diverse citation set — pulling in more companies than the explicitly GEO-labelled queries. This matters for content strategy: pages optimised for "AI visibility" and "AI tracking" queries reach citation in AI responses more reliably than pages optimised only for "GEO" terminology.
Implication for optimisation: Write content for the queries your buyers actually use, not just for the category label you prefer. Buyers searching for "how to track AI citations" and "appear in AI results" are using different language than "GEO optimisation."
Finding 5: Perplexity cites the most diverse set of companies
Comparing citation diversity by engine (how many unique companies receive at least one citation):
- Perplexity: Most diverse — 20 unique companies cited in a typical week
- Gemini: Moderate — 12–15 unique companies
- ChatGPT: Moderate — 12–15 unique companies
- Claude: Most concentrated — 8–10 unique companies
What this means: Perplexity is the highest-opportunity engine for brands trying to break into AI citation for the first time. Claude is the hardest — it retrieves from a narrower set of highly authoritative sources. A practical prioritisation: optimise for Perplexity first, then ChatGPT, then Gemini, then Claude.
Why the gap exists
The 66% with zero citations share a common profile: they operate in the GEO/AEO space, but their own brand has insufficient off-domain citation infrastructure for AI engines to retrieve confidently.
AI engines do not cite brands because those brands have good websites. They cite brands because:
- Third-party sources validate them — directories (G2, Clutch, Capterra), review platforms, industry roundups
- Multiple independent sources agree — not just one listicle, but several independent sources naming the same company
- The domain history and authority of citing sources is strong — a mention in a SearchEngineLand article weighs more than a mention in a new blog
- The brand has been consistently mentioned over time — AI retrieval gives weight to brands that appear repeatedly across sources
Companies that built their domain and off-domain presence before AI engines became dominant are naturally advantaged. Companies entering the GEO space in 2024–2026 need to build this infrastructure deliberately.
What companies in the 66% should do
The off-domain citation gap can be closed systematically. Based on the citation patterns in this study, the highest-leverage moves are:
1. Get listed on the four core review platforms
G2, Clutch, Capterra, and Product Hunt are cited in AI responses across all four engines. These are non-negotiable for any company trying to build AI citation infrastructure.
2. Appear in at least three independent "best of" roundups
AI engines are more likely to cite a brand when multiple independent sources name it in the same category. A single listicle is insufficient; you need consistent repetition across sources.
3. Build a tool or free resource
The 4:1 tool-to-agency citation ratio is stark. A free measurement tool, a calculator, or a data resource gives AI engines a concrete, repeatable reason to cite you in response to tool-related queries — which attract more citation volume than service queries.
4. Prioritise Perplexity-friendly sources
Perplexity cites the most diverse set of companies and appears to weight recently updated content more heavily than other engines. Reddit threads, community discussions, and recently updated industry roundups all index well in Perplexity's retrieval pool.
5. Measure consistently
You cannot manage what you don't measure. Daily citation tracking across all four engines — even at a basic level — shows whether placement work is producing movement and which sources are having the most impact.
About this research
This study was conducted by Kaus AI using a proprietary citation tracking pipeline. All data was collected via direct API queries to AI engine providers. The 82 companies tracked were selected from known GEO/AEO providers and expanded over the tracking period as new entrants appeared in AI engine responses.
This research is updated quarterly. The June 2026 edition covers 14 May to 11 June 2026.
Cite this research: Kaus AI, "The GEO Visibility Gap," June 2026. kaus-ai.com/geo-visibility-gap-report/
Frequently Asked Questions
How was this data collected?
Kaus AI's citation tracking pipeline runs four buyer-intent queries per day per engine across ChatGPT, Perplexity, Gemini, and Claude. Each company mention is logged with date, engine, query, citation position, and context. The pipeline has been running since May 14, 2026.
Why were these 82 companies selected?
The tracked companies include all known GEO/AEO agencies and tools identified as of May 2026, plus new entrants that appeared in AI engine responses during the tracking period. The list is expanded continuously as new companies enter the space.
What counts as an AI citation?
A citation is recorded when a company name appears in an AI engine response to one of the four tracked buyer-intent queries. Position (first, second, or later mention) is recorded separately from mention/no-mention. The 66% finding refers to any mention — even companies appearing once count as cited.
Does this mean 66% of GEO agencies are bad?
Not necessarily. Many excellent agencies serve clients well without having high AI citation frequency for their own brand — especially newer agencies or those working primarily with referrals. The point is that AI citation frequency is an increasingly important signal as buyers use AI engines for vendor discovery. Agencies with zero citations risk being excluded from the consideration set of buyers who start with an AI engine question.
How can I find out where Kaus AI sits in this data?
Kaus AI is a new brand (launched 2026) and was in the zero-citation group at the start of the tracking period. This research was motivated in part by our own experience building AI citation infrastructure for a new brand. The methodology we document is what we apply for our clients.
Kaus AI is a London-based GEO (Generative Engine Optimisation) agency. We measure and build AI citation infrastructure for SMBs across the UK and Europe. kaus-ai.com