One assumption underpins most current approaches to AI search visibility, and the data does not support it. The assumption is that ranking well in conventional search produces citations in AI answers, and that existing search investment therefore covers the new channel automatically.
Royston G King argues that this assumption is the most common reason established brands find themselves absent from AI answers despite years of search investment. A Forbes 30 Under 30 Monaco honouree, University of Southern California alumnus accepted into Columbia University, and contributor to Entrepreneur, he founded Master Scaling in 2018 and Quantum Scaling Partners as its selective arm, and has spent close to a decade watching the two systems diverge.
Studies examining large sets of prompts have repeatedly found weak overlap. A substantial share of pages cited by AI assistants do not rank in the top hundred conventional results for the same query. Analysis of the pages ChatGPT cites most frequently has found a meaningful portion with negligible conventional organic visibility. The relationship exists but is far looser than most teams assume.
Understanding why matters, because the reason determines what to do about it.
Conventional ranking is a competitive ordering. Pages are scored against one another and placed in sequence, and the ordering is heavily influenced by link signals, engagement, and site-level authority accumulated over years. AI citation works differently. A system assembles an answer from a pool of material it considers relevant, then selects passages that support the specific claims it is making. Selection favours passages that state something clearly and specifically, from sources the system treats as reasonable, in language that matches the question.
A page can be an excellent answer to a narrow question while ranking poorly because it lacks the accumulated authority signals that determine competitive placement. Conversely, a page can rank first through domain strength while containing nothing directly quotable for a specific query.
This produces several practical consequences.
Conventional rank tracking will systematically miss AI performance. A brand can hold strong rankings and remain absent from AI answers, or hold weak rankings and appear frequently. Reporting built entirely on position tables provides no visibility into the second channel at all.
Impressions correlate better than clicks. Analysis has found the relationship between conventional search performance and AI citation strengthening considerably when measured against impressions rather than clicks. The reason is mechanical. AI systems draw from the pool of pages deemed relevant to a query, which is what impressions describe. Clicks describe which page won a competition that AI citation does not run.
Domain authority matters less than topical fit. A trade publication with modest domain metrics covering a specific industry frequently outperforms a large general interest outlet for specialised queries, because the surrounding content establishes topical relevance. Budget directed at securing the largest available logo is often worse spent than budget directed at consistent coverage in the places a category is genuinely discussed.
Content structure matters more than it does for ranking. Material written as a continuous argument, with the conclusion arriving at the end, offers little that is extractable. Material that states a claim plainly, then supports it, offers a passage a system can lift. This is a writing decision rather than a technical one, and it is largely independent of anything that affects conventional ranking.
None of this argues for abandoning conventional search work. Organic search remains the larger channel by volume for most organisations, and the technical fundamentals of crawlability and site structure serve both systems. The argument is against assuming coverage. An organisation that has invested substantially in search for years may have built very little of what determines AI citation, because the two systems reward different things.
The practical starting point is measurement. Running a defined set of prompts across major assistants at intervals, recording whether the brand appears and which source was used, produces a picture that no ranking report contains. Most organisations conducting that exercise for the first time find the results surprising in both directions.
Most organisations running that measurement exercise for the first time, King reports, find the results surprising in both directions. Quantum Scaling Partners begins client engagements with it for exactly that reason.
Connect With Royston G. King
To learn more about Royston G. King and follow his latest work, visit his official website, connect with him on Instagram and LinkedIn, or watch his latest content on YouTube.
