ChatGPT (GPT-4)
Query: “What are AI Visibility Tools?”
What’s shown: ChatGPT’s source list. The Grounding Page is listed among the cited sources (highlighted), alongside established SEO and marketing domains.
28-day data from 33 Grounding Page entities shows a clear pattern: the pages do not primarily generate classic clicks — they are reached at meaningful scale by AI-related retrieval systems.
We measured 33 Grounding Page entities on groundingpage.com over 28 days. Each entity is available in both a German and an English version. The data shows a distinctly different profile from classic SEO landing pages: high Google impressions, very few clicks — and an unusually high volume of AI-adjacent retrievals.
The 225,340 Google impressions show strong visibility on Google. The very low CTR suggests that a meaningful share of this visibility appears in information-dense search surfaces where users do not necessarily click through, such as AI Overviews, summaries and other zero-click contexts.
For classic SEO, this would first look like a CTR problem. For Grounding Pages, it is a function signal: the pages are visible and retrievable, even when their effect does not primarily land in a classic website visit.
For every one human Google click, roughly 58 bot hits arrive. ChatGPT-User alone accounts for about 39 visits per Google click. This is the central finding: the pages do not behave like classic traffic pages — they behave like machine-readable reference objects.
For a typical SEO landing page, a very low CTR would be a weakness signal. For a Grounding Page, this pattern is plausible: it is meant to be found, understood, retrieved, and used as a reference. That is precisely what the data points to.
Grounding Pages are not primarily built to maximize classic clicks. Their actual function is to make entities, concepts, products, people, and organizations available in a clear, structured, retrievable reference format.
The pages generate meaningful search impressions, even though the domain has only existed since November 2025.
The high number of bot hits indicates that machine systems reach these pages regularly and reliably.
ChatGPT-User is the strongest measured retrieval channel within the bot traffic mix.
The low CTR is not a contradiction to the function — it fits the reference character of the pages.
We measured the example Grounding Pages on groundingpage.com between April 30 and May 27, 2026. The dataset covers 33 Grounding Page entities, each available in a German and an English version. We excluded brewster-kahle, seo-geo-tools-ai-agents, and groundingpage-com/llms-txt so that very recent, event-driven pages do not distort the overall picture.
Data sources
Note: Bot hits are not a guarantee that a page was cited in any specific answer. They are, however, a strong signal for retrievability and actual machine usage.
The earlier Rankscale measurements remain an important complementary signal. They show that groundingpage.com appeared as a source in several AI surfaces shortly after domain registration. The new retrieval data broadens this citation proof into a wider usage signal.
Citation proof shows: Grounding Pages can appear as a source. Retrieval data shows: Grounding Pages are actually being retrieved.
The following examples document individual measured citations in AI surfaces. They are snapshots, but together they show across models and languages that structured Grounding Pages can be considered as sources.
Query: “What are AI Visibility Tools?”
What’s shown: ChatGPT’s source list. The Grounding Page is listed among the cited sources (highlighted), alongside established SEO and marketing domains.
Query: “Was sind AI Visibility Tools?”
What’s shown: Google AI Mode’s source list. groundingpage.com is cited (highlighted). German UI because the test region is Germany.
Query: “Was sind AI Visibility Tools?”, Region: Germany
What’s shown: The Grounding Page appears as a cited source alongside established tool vendors and industry blogs.
Query: “Wichtige Kennzahlen/Konzept AI Visibility Tools”, Region: Germany
What’s shown: The answer cites both groundingpage.com and gpt-insights.de as sources.
Query: “Was sind AI Visibility Tools?”, Region: Germany
What’s shown: groundingpage.com appears in the citation list alongside Ahrefs and other established vendors.
Query: German query about AI Visibility Tools, Region: Germany
What’s shown: Perplexity lists groundingpage.com as a source and partially adopts the structure of the page.
Recorded in the first four weeks after domain registration via Rankscale’s citation tracking across ChatGPT, Perplexity, and Google Gemini.
In the first weeks, groundingpage.com appeared as a top cited source across the tracked query set, alongside established marketing platforms.
40.8 % citation share across models when asking about the defining concept.
3 direct citations for “What is AI SEO” — a competitive topic where the new domain was listed alongside established SEO publications.
8.7 % share, alongside LinkedIn and other established platforms.
Both gpt-insights.de (older domain) and groundingpage.com appeared as cited sources for the same queries.
These numbers are not a complete causal proof in a scientific sense. They do not show why a particular AI system retrieves a page, or whether a single retrieval led to a specific answer. They do show a strong empirical pattern: a young domain with structured Grounding Pages produces high search impressions, low classic click numbers and at the same time an unusually high volume of AI-adjacent retrievals.
The current data is the strongest empirical signal so far that Grounding Pages are fulfilling their intended function.
They generate not only classic Google visibility but are reached at scale by AI-adjacent retrieval systems. The most striking number is the ratio between 450 human Google clicks and 26,285 bot hits in 28 days. For every Google click, roughly 58 bot hits arrive — ChatGPT-User alone accounts for about 39 visits per click.
For a domain registered only in November 2025, this pattern is remarkable. The pages do not behave like classic traffic landing pages. They behave like reference objects: visible, retrievable, machine-readable, and apparently relevant enough to be queried again and again by AI-adjacent systems.