AI Search Visibility Keynote: From Invisible to Cited, Inside 52,821 AI Citations
AI search visibility is whether an AI assistant cites a person or company when someone asks it a question that needs an answer of that kind. It decides who gets found, booked, hired or shortlisted, because the person asking often never opens a search results page at all.
Gartner projects traditional search engine volume will fall 25% by 2026 as AI chatbots absorb queries that used to land on Google [1]. A keynote audience already accepts that shift. What they want is a working model of why AI systems cite what they cite, backed by evidence rather than opinion. This article covers what such a keynote contains, which audiences it fits, and the study behind it.
Why Does AI Search Visibility Now Decide Who Gets Booked?
Being cited by an AI assistant increasingly decides who gets contacted first, because the person asking stops looking once the assistant names someone. The shortlist is formed before any human comparison begins.
Adobe Analytics recorded AI-chatbot referral traffic to US retail sites surging during the 2024 holiday season, evidence that people already let an assistant's answer replace their own browsing [2]. The same behavior applies to conference organizers scouting speakers, procurement teams shortlisting vendors and candidates researching employers. For a speaker, this changes what "known" means. A strong LinkedIn following or a well-optimized website used to be enough to get found. Now, when an event organizer asks an assistant "who should we book for a keynote on AI and search," the answer comes from whatever the model has indexed as citable, and audience size does not enter into it.
What Does the 52,821-Citation Study Show About Getting Cited?
A pre-registered study of 52,821 AI citations found that matching the query's meaning beats every authority and formatting signal combined, backlinks included. It tracked 1,362 queries across ChatGPT and Google AI Overviews over three snapshots in July 2026, modeling 20 candidate ranking factors from the published GEO literature and reaching a discrimination accuracy (AUC) of 0.86 [6].
Three findings carry directly into a boardroom or a keynote hall. Domain authority and backlinks lost statistical significance once page content was accounted for (p=.89 and p=.94), so a well-linked page with a weak answer loses to a thin page with a strong one. FAQ structured data lifted citation odds by 43%, a real and measurable formatting effect. And Google AI Overviews penalize owned brand pages (std-OR -0.73, p=7e-06) while ChatGPT does not, which makes ChatGPT visibility and AI Overview visibility two separate fights. The full breakdown lives in the GEO and AEO data guide.
Why Do Most AI Visibility Speakers Talk Theory Instead of Data?
Most speakers on this topic repeat secondhand claims because they are reselling a tool vendor's marketing deck rather than running their own experiments. The field is young enough that rigorous, pre-registered work on what drives citation is scarce.
The consequence shows up on stage as plausible-sounding tactics: post more, add keywords, chase backlinks. Two of those three did not survive contact with the citation data. A vendor-affiliated speaker also carries an incentive to recommend whatever the sponsoring tool automates, whether or not the numbers support it. An independent speaker running the underlying experiment can say plainly which popular tactic failed, and show the p-value next to it. Ahrefs found only 12% of AI-cited URLs also rank in Google's top 10 [3], a result that quietly invalidates a large share of standard SEO advice for this specific problem.
What Does an AI Search Visibility Keynote Actually Cover?
The keynote walks an audience through what predicts AI citation, using the live model and real numbers. It opens with the shift itself, then moves into the findings from the 52,821-citation study.
The core segment covers relevance beating authority, FAQ schema working where question-form headings alone do not, and the two dominant AI engines rewarding opposite content strategies. Legacy ranking-factor work from Backlinko's 11.8M-page study [4] and the Semrush ranking factors study [5] serves as the comparison baseline, which is what makes the divergence legible. The second half turns the data into a plan a marketing, sales or communications team can act on inside a quarter: what to rewrite first, what to stop spending on, and how to check citation status without expensive tooling.
Want the data on your stage? I run the AI-citation experiments behind this talk myself, then present what the results mean for marketing, sales and comms teams. Book a 15-minute intro call → to check dates and fit.
How Is This Keynote Different From a Generic AI Trends Talk?
Every claim in this keynote traces to a named number from a pre-registered study, which makes each one falsifiable on stage. A generic AI trends talk bundles search visibility with productivity tools, agents and automation into one broad narrative, useful for awareness and thin on anything an audience can apply the next morning.
This talk stays narrow on purpose. It covers one mechanism, what makes an AI system cite a page or a person, tested against 20 candidate variables and reported with the failures included. The myth-scoreboard format, common claim versus what the data showed, gives the audience a structure they can repeat internally and gives skeptical attendees a reason to trust the parts that are new to them. The same comparison method applied to a full site audit is documented in the walkthrough of going from invisible to recommended by AI.
What Sets a Data-Backed Speaker Apart From a Tool-Vendor Speaker?
A keynote audience buys a mental model that outlives any one tool's feature set, which is where the two speaker types diverge most sharply. The distinction matters most when an organization decides its own AI visibility strategy in the weeks after the event.
| Criterion | Tool-vendor AI visibility speaker | Data-backed AI visibility speaker |
|---|---|---|
| Recommendations | Tactics the sponsoring tool automates | What the citation data supports, tool-agnostic |
| Evidence | Secondhand industry claims and case studies | A named, pre-registered study with the method shown |
| Update cycle | Talk changes with the product roadmap | Talk updates as new citation snapshots are analyzed |
| Transferability | Framework proprietary to one platform | Findings apply across ChatGPT, AI Overviews and beyond |
The table names the four checks an event organizer can run on any AI visibility speaker before signing.
Which Audiences Get the Most Value From This Talk?
Marketing and communications leadership teams get the most direct value, because AI search visibility sits inside their existing content and brand mandate. In practice the audience runs wider than that.
Sales leadership benefits because a prospect's AI assistant shapes the shortlist before a rep gets a call. HR and employer-branding teams benefit because candidates ask assistants which employers to trust before applying. The strongest bookings so far have been marketing and communications conferences, in-house leadership offsites with AI strategy on the agenda, and industry events in categories such as finance, manufacturing and professional services, where a competitor being cited while the company is not is a live commercial risk. A related talk on synthetic audience testing is documented under digital twins keynote speaker.
What Results Can an Audience Expect From This Keynote?
An audience leaves with a working checklist: which content changes move AI citation odds, which classic SEO habits waste budget on this specific problem, and how to check their own visibility gap in the room using a live example.
That testable takeaway is why the format works as both a 45-minute conference keynote and an internal half-day workshop, since attendees can act on it before the next quarterly planning cycle. Post-talk requests have run toward a follow-up audit of the organization's own AI citation status using the same method behind the study, and a workshop version where a marketing or content team rewrites two or three real pages during the session. Both extend from the keynote without a separate sales pitch.
Conclusion: Citation Status Is a Commercial Metric, Not an SEO Detail
AI search visibility decides who appears in the first AI answer, with measurable effects on inbound requests, leads and bookings. A talk on this topic holds up only when it rests on audited citation data: 52,821 citations, 1,362 queries, an AUC of 0.86 and the failed hypotheses reported alongside the confirmed ones.
Planning an agenda around AI visibility? I turn this 52,821-citation study into a keynote or workshop, in English or German, with the live model and the myth scoreboard on screen. Schedule your intro call →
Sources
- Gartner (2024). Gartner Predicts Search Engine Volume Will Drop 25% by 2026, Due to AI Chatbots and Other Virtual Agents. gartner.com
- Adobe (2024). Adobe Analytics: AI Chatbot Traffic to US Retail Sites Surged During 2024 Holiday Season. news.adobe.com
- Ahrefs (2024). Only 12% of AI Cited URLs Rank in Google's Top 10. ahrefs.com
- Backlinko. Search Engine Ranking Factors (11.8M-page study). backlinko.com
- Semrush. Ranking Factors Study. semrush.com
- Mall, J. (2026). GEO and AEO Ranking-Factor Study: 52,821 AI Citations. jonathanmall.com. GEO and AEO data guide
About the author: Dr. Jonathan Mall is a cognitive neuropsychologist (PhD, University of Groningen, 2013), CIO and co-founder of neuroflash, and a keynote speaker on AI search visibility, GEO and consumer psychology. He runs the ranking-factor experiments behind this talk himself on the Rankscale citation panel for jonathanmall.com, and presents the results in English and German across the DACH keynote circuit. Facts, Q&A and press kit · LinkedIn
Frequently Asked Questions About AI Search Visibility Keynotes
What is AI search visibility?
AI search visibility is how often, and how favorably, a person, brand or company gets cited by AI systems such as ChatGPT, Google AI Overviews and Perplexity when someone asks a question that needs that entity in the answer. It replaces the older idea of a search ranking with a simpler test: does the model quote you, or quote a competitor. An analysis of 52,821 AI citations found that being the best-matching answer to the query predicts citation far more than backlinks, domain age or page authority.
Why should a conference or company book a keynote speaker on AI search visibility?
The topic reaches well past marketing. Sales teams lose deals to a competitor an AI assistant recommended first, HR teams get evaluated by candidates who asked an AI which employer to trust, and executives make budget calls based on what an assistant told them about a market. A keynote on AI search visibility gives all of those functions one evidence-based model of why AI systems cite what they cite, in place of five different guesses in five departments.
What makes this keynote different from a typical AI trends talk?
The speaker ran the underlying study. Jonathan Mall built a pre-registered ranking-factor model over 52,821 real AI citations across 1,362 queries and two engines, and presents the live results, including the claims that did not hold up. The talk names what moves citation odds and what does not, with the statistics on screen.
Does AI search visibility replace traditional SEO?
No, it sits alongside it and runs on partly different rules. Ahrefs found only 12% of AI-cited URLs also rank in Google's top 10, so a page can win in one system and stay invisible in the other. The citation data adds a further split: Google AI Overviews penalizes owned brand pages while ChatGPT does not, so one SEO strategy cannot serve both answer engines identically.
Who is Jonathan Mall and why does his data carry weight?
Jonathan Mall is a cognitive neuropsychologist (PhD, University of Groningen) and co-founder of neuroflash who runs AI-citation experiments on his own site's Rankscale panel instead of citing secondhand industry claims. The keynote is built from a pre-registered 20-variable study reaching an AUC of 0.86, with the method, the effect sizes and the failed hypotheses published.
How can a company measure its own AI search visibility before booking a talk?
Ask an AI assistant the exact questions a buyer or candidate would ask about the company, its category and its named competitors, then note who gets cited and who does not. That manual check surfaces the gap fast; a full audit adds citation tracking across engines over time. The keynote covers the do-it-yourself check and what a structured audit adds once the gap is confirmed.