Aemorph's cluster-level measurement shows how often AI answers name your business across the questions buyers ask inside one buying decision, not where one page ranks for one phrase. It reports the share of tested answers within a defined prompt cluster, a group of related buyer questions, that mention, cite or recommend you. A buyer researching one decision asks several related questions across several sessions, so you need to show up across that cluster, not on one keyword. Clear, accurate brand information supports a mention; inclusion also depends on the question, the sources retrieved and the engine answering.
Written by Kevin Dam, founder of Aemorph, who has practised SEO since 2010 and sits on the Forbes Agency Council. Aemorph is a Google Partner. LinkedIn
What Single-Keyword Tracking Misses
A first-place ranking for one keyword does not show whether buyers meet your business across the other questions they ask before deciding, so a ranking report leaves out the research where your next buyers form their shortlist. A single keyword ranking tells you how one page performs for one phrase. It says nothing about whether ChatGPT, Claude, Gemini or Perplexity names your business when a buyer asks any of the other related questions inside the same decision.
A buyer choosing a discoverability consultancy does not ask one question, and often spreads the questions across several separate AI chat sessions. Inside one decision, the buyer:
- Checks what the service costs.
- Asks how long it takes.
- Compares it with standard search optimisation to see how it differs.
- Matches it against their own industry.
- Confirms it works for their jurisdiction.
You win the single tracked keyword and still have no reading on the citations that matter commercially, because those sit in the questions nobody was tracking. Pew Research's click data shows why a click-based number is an incomplete measure: an AI summary appeared on 18% of US Google searches in March 2025; 58% of users in the study encountered at least one AI summary (Pew Research, 2025), and on the pages where a summary appeared, users clicked through to a traditional result link in only 8% of visits, against 15% without one (Pew Research, 2025). Reporting built only on blue-link clicks leaves out the research that ends inside the summary itself.
Users clicked a source cited inside the summary in only 1% of visits.
| Search page | What the user clicked | Share of visits | Sample | Source |
|---|---|---|---|---|
| No AI summary | A traditional result link | 15% | 900 US adults, March 2025 | Pew Research Center, 22 Jul 2025 |
| With an AI summary | A traditional result link | 8% | 900 US adults, March 2025 | Pew Research Center, 22 Jul 2025 |
| With an AI summary | A source cited inside the summary | 1% | 900 US adults, March 2025 | Pew Research Center, 22 Jul 2025 |
Source: Pew Research Center, "Google users are less likely to click on links when an AI summary appears in the results", July 2025.
Why does one keyword ranking not tell the whole story?
One keyword ranking does not tell the whole story because a buying decision is made across a cluster of related questions, and an AI assistant citing your business for one of them does not mean it cites you for the rest. Coverage across the cluster, not a position on one phrase, is the better indicator of whether a buyer's overall research names your business or a competitor's.
How Cluster-Level Measurement Works
Aemorph groups the related questions inside a buying decision into one defined cluster, then measures what share of the tested answers mention, cite or recommend your business, so you see where buyers meet you and where a competitor is named instead. The starting point is the buyer's decision itself, not a keyword list. AI keeps a map of what every business is and what it does. Each engine answers from what it has learned and looked up about you across many sources, and your website is only one of them.
The measurement runs in four steps:
- Start from the decision. Aemorph names the buying decision a buyer is researching, such as choosing an AI visibility consultancy.
- Map the questions. Aemorph maps the questions a buyer plausibly asks while researching that decision into a defined panel.
- Group the questions into one cluster. Aemorph groups those questions into one cluster and treats the cluster as the unit that gets measured.
- Measure the cluster across engines. Aemorph checks each question in the panel on each engine tracked and reports the share of tested answers that name your business.
For each cluster, the measurement checks three things: whether your business gets mentioned at all, whether it gets recommended rather than just mentioned in passing, and whether that result holds across the four AI chats (ChatGPT, Claude, Gemini and Perplexity) plus Google AI Overviews, or drops out on specific engines. GEO (generative engine optimisation) covers the four AI chats, where the measure is citation inside the chat answer. AEO (answer engine optimisation) covers Google AI Overviews, featured snippets, answer boxes and voice search answers; this cluster report checks Google AI Overviews. SEO (search engine optimisation) covers Google's ranked results, which your ranking report tracks.
Cluster-level measurement reports on the coverage built through Aemorph's attribute mapping methodology and the other methods inside the LLM Influencer System, Aemorph's SEO, AEO and GEO programme.
| Metric | What counts | Engines and answer surfaces checked | Example | Why it matters |
|---|---|---|---|---|
| Mentioned | Your business named anywhere in the answer | The four AI chats and Google AI Overviews | ChatGPT named Aemorph as a source for "best PSG SEO services" in a logged-out session on 4 August 2026 | A business the answer does not name is absent from the shortlist that answer gives the buyer |
| Recommended | Your business put forward as an option the buyer acts on, not named only in passing | The four AI chats and Google AI Overviews | Report line in the form "Recommended, not only mentioned, in ChatGPT's answers on cost and timing" | A passing mention puts you in the answer; a recommendation puts you on the buyer's shortlist |
| Consistent | The same result on each engine and surface tracked | The four AI chats and Google AI Overviews | Report line in the form "Named on ChatGPT, Claude and Perplexity; still missing on Gemini and Google AI Overviews" | A buyer who switches engines mid-research meets a different answer |
Source: Mentioned example from a logged-out ChatGPT session on 4 August 2026, query "best PSG SEO services".
| Unit | What it tracks | Engines and surfaces covered | Example report line | What it misses |
|---|---|---|---|---|
| Single keyword ranking | One page's position for one search phrase | One search engine's ranked results | In the form "Position 4 for 'discoverability consultancy singapore'" | The other questions inside the same buying decision, and the AI answers |
| Topic cluster share | Coverage across the defined panel of related questions | The four AI chats and Google AI Overviews | In the form "Cited in ChatGPT's answers to 7 of 10 tracked questions, up from 4 of 10", one line per engine | Questions and wordings outside the defined panel, and answer variation between sessions |
A cluster report showing coverage across eight of ten related questions, with two clear gaps, gives Aemorph a specific, fixable target: the two questions where your business is missing. A single-keyword report showing one ranking gives no such map of where the gaps sit.
What This Reporting Looks Like in Practice
A cluster report shows your share of answer coverage across a named buying decision, tracked over time, so you see whether more of your buyers' research names you, not a list of keyword positions moving up and down. Instead of "position 4 for 'discoverability consultancy singapore'", each line of your cluster report reads in the form "cited in ChatGPT's answers to 7 of 10 tracked questions inside the 'choosing an AI visibility consultancy' cluster, up from 4 of 10 the prior period, with no citation yet in Gemini or Google AI Overviews for the same questions."
This framing matches how GEO and AEO visibility behaves for a buyer: closer to how consistently a brand's reputation shows up across many separate conversations than to how one page performs in one search result. Your monthly reports group results by topic and prompt clusters, alongside ranking reports: the cluster view maps to buyer behaviour, and the ranking report keeps the search engine's own measure in view. One combined report covers a website that serves both markets, and separate reports cover separate sites, for example a .com.sg and a .com.au.
How is this different from a standard rank-tracking report?
A standard rank-tracking report lists keyword positions for individual pages on one search engine; a cluster report lists the share of a defined question set in which your business is named across the AI engines tracked. It answers the commercial question a rank-tracking report leaves open: how often a buyer researching your category across several questions and several AI tools meets your business in the answers tested, or a competitor's. You receive both: results grouped by topic and prompt clusters, and ranking reports.
Put This to Work on Your Reporting
Start with the Showstopper Sensor (Sprint 1), Aemorph's 30-day AI answer sprint, which produces the answers to your showstopper questions for your Brand Codex by day 30. Showstopper questions are the buyer questions that stop a purchase when AI answers them badly, and your Brand Codex is your brand fact sheet: the single source of truth for what AI says about you. Cluster-level measurement is how Aemorph measures visibility across the LLM Influencer System.
Buy the Showstopper Sensor (Sprint 1) on its own and it carries the Ground Truth Guarantee: if ChatGPT, Claude, Gemini and Perplexity still miss your approved facts 30 days after your answers go live, you get your Showstopper Sensor fee back in full.
A strategic conversation with Aemorph is free and runs 30 to 60 minutes, and Aemorph responds within 24 hours to all communication.