In AI Search, Authority Gets You Cited. Desire Gets You Chosen.

TL;DR

• Authority gets you into the answer. It does not guarantee you are the choice. AI answers narrow a market down to a handful of credible options, and choosing between them is rarely a purely rational act.

• Beyond its offer, a brand becomes desirable through an idea, an emotion, a vision and an experience. Decades of effectiveness research, and Europe's marketing leaders today, point in the same direction.

• But the perfect world of advertising is increasingly exposed. Engines combine what a brand says with what others say about it, and buyers check too. A promise nothing corroborates stays marketing.

• Desire does not enter an AI answer by magic. It travels through what third parties write, through buyers who come looking for the brand by name, and through the choice made after the answer.

• The rule we apply at PingPrime: dream big, never make things up.

Authority gets you cited. It does not guarantee you are chosen.

This article started with a comment. Under a LinkedIn post on the Amazon and OpenAI advertising partnership, a reader asked a question I had not asked myself: if AI becomes the rational, factual voice that compares and recommends, what will still make people want one brand rather than another?

It is a fair challenge to our own discipline. Most of what we write about GEO is about authority: being cited, being described correctly, being corroborated by independent sources. That work is necessary. But look at what an AI answer does. It narrows a market down to a handful of credible options, and how far it goes varies by engine. In a study of about 10,000 URLs cited by seven AI surfaces, Scrunch found that when an answer cited an independent comparison page, it went on to recommend a specific brand in 34% of ChatGPT responses, 50% of Claude responses and 64% of Gemini responses. The rest of the time, the final choice stays with the buyer.

And when several options are credible, reason alone rarely settles the choice. What tips it is what a comparison cannot produce on its own: a point of view, a feeling, a future people want to be part of. Desire.

That is also why we say GEO is branding. Not because GEO replaces the brand, but because it puts the brand to the test.

What makes a brand desirable beyond its offer?

Products and services can be compared line by line, and AI engines are very good at that comparison. What they cannot manufacture is the reason to care. We find it useful to think of that reason in four levers. Each one has an authentic version and a fictional one, and the difference matters more than ever.

PingPrime framework.

Being authentic does not mean being bland. Being realistic does not mean giving up on imagination. A brand is allowed to aim high and to open a horizon. The condition is simple: every dream has to rest on something real.

Why emotion is not the enemy of evidence

Some GEO practitioners treat emotion as noise: engines want facts, so give them facts. That reading ignores a large body of marketing evidence. In The Long and the Short of It, Les Binet and Peter Field analysed close to a thousand campaigns entered into the IPA Effectiveness Awards between 1980 and 2010. Emotionally driven work produced more brand effects and more business effects than rational messaging, and those effects lasted longer. The scope matters: these are advertising campaigns, mostly consumer-facing, selected because they were strong enough to compete for an award, and all predate generative AI. The lesson we take from them is about how people choose, not about media.

European marketing leaders are drawing a similar conclusion today, and not only in consumer markets. McKinsey's State of Marketing Europe 2026 surveyed 500 senior marketing leaders in France, Germany, Italy, Spain and the UK, 44% of them in B2B. Branding ranks first out of 20 topics and authenticity fourth, and about 70% of respondents say purpose-driven, authentic brand experiences are essential to build distinctive emotional connections. The report's conclusion is close to our thesis: as algorithms take over tasks like product comparison, McKinsey expects emotional attachment and authenticity to become the primary drivers of brand value.

In the same report, Richella Odebrecht, who leads global marketing for industrial automation at Schneider Electric, describes a B2B campaign built to create a feeling, a desire to be that person, rather than just deliver a message. Emotion and evidence are not opposites, in B2C or in B2B.

Why the invented world is increasingly exposed

Advertising has trained us to expect perfect worlds: customers who are always smiling, promises without friction, lives transformed by a product. In a channel where the brand controlled the message, that fiction could work for a while.

Generative search changes the conditions, though not absolutely. Engines do reuse what brands publish about themselves. But when they answer a question about a brand, they also draw on what the press, reviewers, experts and customers say. Graham Kenny and Ganna Pogrebna, writing in Harvard Business Review, describe branding as moving from paid persuasion to AI-mediated recommendation, with third-party validation and consistent positioning now shaping visibility. And the check is not only algorithmic: buyers who meet a brand in an answer go on to read reviews and ask peers.

The consequence for desire is direct. A promise that nothing corroborates stays marketing. A vision anchored in evidence can become a reference. Andy Crestodina of Orbit Media makes the practical case in the Content Marketing Institute's 2026 trends roundup: publish the proof points that win the sales conversation (case studies, impact metrics, reviews, testimonials) on your site and anywhere else you can, because without published reasons to believe, AI is likely to recommend a competitor.

The sameness trap: what is generic gets summarized

There is a second reason authenticity matters. Many brands now produce content with the same generative tools, fed with similar prompts. An experiment published in Science Advances by Anil Doshi (UCL) and Oliver Hauser (University of Exeter) illustrates the risk in a different field: writers who used AI-generated ideas produced short stories judged more creative individually, but those stories were more similar to one another than stories written without AI. The study is about fiction, not brands, but the dynamic is easy to recognise in marketing: better on average, less diverse as a whole.

For an engine building an answer, interchangeable brands offer little to say beyond a list of attributes. Kenny and Pogrebna make the positive version of the point in HBR: firms should make sure their ideas, data and distinctive concepts are clearly associated with the brand, by publishing original data and naming proprietary frameworks.

Hence a conviction we hold at PingPrime, which we offer as a working principle rather than a measured law: what is generic gets summarized; what is singular gets cited. A brand with a clear idea, real stories and a recognizable voice gives the engine something to say. A brand without them risks being reduced to a line in a comparison table.

Experience: the lever Google already named

Of the four levers, experience is the one search engines formalized first. In December 2022, Google added "Experience" to its quality rater guidelines, turning E-A-T into E-E-A-T: for many topics, what people value most is content produced by someone with first-hand, lived experience of the subject. Two limits apply. These guidelines are used to evaluate Google's ranking systems and do not directly influence rankings. And the framework is Google's own: ChatGPT, Perplexity and other assistants do not publish equivalent guidelines. The signals it describes travel, though. First-hand detail, identifiable authors and a reputation confirmed by other sources are exactly what any engine can find when it retrieves and weighs sources, which is why E-E-A-T has become a useful checklist well beyond Google.

For a brand, experience works on both sides of the signal. On-site, it means content built from real cases, real customers and real teams rather than generic claims. Off-site, it means the experience people live with the brand and then tell elsewhere: in reviews, in communities, to journalists. That second part is what turns a brand's story into corroborated authority, the kind of signal engines can find and cross-check.

Practitioners converge on the same point. In the same CMI roundup, Farra Kober of BBC StoryWorks argues that resonance will come from the human stories brands choose to tell, and Ingrid Booth of Investec advises brands to offer what AI cannot: genuine voices, fresh insights and emotional connection.

How desire actually reaches an AI answer

An engine does not feel anything. So how can desire, something emotional, matter in AI search at all? Not directly. It travels through three paths, which correspond to layers of how we measure narrative authority:

1. Through what others write. When customers are moved by a brand, they say so in reviews, testimonials, forums and press coverage. Those are the third-party sources engines draw on. This is the Source share layer: which independent sources feed the answers about you, and what they convey.

2. Through buyers who come looking for you. A brand people remember gets searched by name. That shows up in branded search and direct visits, the Recall layer, and in the questions people ask engines about the brand specifically.

3. After the answer. Once the engine has narrowed the options, the buyer usually still chooses. Desire does its most visible work at that moment, and it shows up in the pipeline: the Revenue layer. As assistants start to shop and act on the user's behalf, that window may narrow, which moves desire further upstream: into the brands people name in their own prompts.

This is why desire and authority are not two separate projects. The same true stories that make a brand desirable to people are the ones that give engines independent material to cite.

How to build content that is both desirable and verifiable

Concretely, here is how we approach it with the brands we support:

1. Write down the idea you defend. Not a tagline: a point of view your leadership would argue for in public. Make it part of your canonical sentence and repeat its attributes consistently across your site, your profiles and your interviews.

2. Collect real stories, with their numbers and their imperfections. A customer case with a measurable result and an honest obstacle is more desirable, and more citable, than a flawless testimonial.

3. Let others tell the experience. Encourage reviews, give customers and employees a voice, open your doors to journalists. Desire that exists only on your own site is not corroborated.

4. Anchor the vision in what you are really building. Share the roadmap, the commitments and the investments that make your ambition credible. Ambition is welcome; fiction is not.

5. Check what the engines retain. Ask ChatGPT, Perplexity and Google's AI features to describe your brand, and look at which sources they cite. If your idea, your vision and your customers' experience are absent, that is the gap to close.

Dream big. Never make things up.

AI engines are becoming a place where people search, compare and get recommendations. That does not make creativity obsolete. It raises the bar for it. The brands that stand out will not be the ones with the most polished promises, but the ones whose promises hold up when an engine, and a buyer, check them against the rest of the web.

The question worth asking isn't "how do we make our brand more attractive?"

It's "what, in our brand, is both desirable and verifiable?"

FAQ

What makes a brand desirable in AI search?

Beyond its products and services, a brand becomes desirable through four levers: an idea it defends, an emotion its customers genuinely feel, a vision anchored in where it is really going, and an experience people live and tell elsewhere. In AI search, each lever carries more weight when independent sources confirm it.

Is emotional content useful for GEO, or do AI engines only want facts?

Both matter. Engines rely on verifiable information to decide whom to cite, but people still choose between credible options on more than reason. Binet and Field's analysis of IPA Effectiveness Awards campaigns found that emotionally driven advertising produced stronger and longer-lasting brand and business effects than rational messaging. For GEO, the key is emotion rooted in real, documented experiences.

How does desire influence AI answers if engines have no emotions?

Indirectly, through three paths: customers who express what they felt in reviews, forums and press coverage that engines cite; buyers who remember the brand and search for it by name; and the final choice buyers make after the engine has narrowed the options.

Can AI engines tell when a brand exaggerates?

They do not judge sincerity, and they do reuse brand-published content. But they also draw on what other sources say. A claim that no review, article or expert confirms has little independent support, which makes it weaker than a claim corroborated across several sources, and buyers can check it too.

How can a brand stay authentic without becoming bland?

By separating ambition from fiction. A brand can share a bold vision, strong convictions and an inspiring future, as long as each rests on something real: commitments, investments, customer results. Authenticity removes invented worlds, not imagination.

Sources

• Les Binet and Peter Field, The Long and the Short of It, IPA, 2013; IPA, "The evidence that emotionally appealing advertising is more effective".

• McKinsey & Company, "Past forward: The modern rethinking of marketing's core" (State of Marketing Europe 2026), including the interview with Richella Odebrecht, Schneider Electric.

• Graham Kenny and Ganna Pogrebna, "LLMs Are Overtaking Search. Here's How to Adjust Your Online Presence", Harvard Business Review, 6 March 2026.

• Content Marketing Institute, "42 Experts Name the Most Important Content Marketing Trends for 2026", 9 December 2025 (contributions by Andy Crestodina, Farra Kober and Ingrid Booth).

• Anil R. Doshi and Oliver P. Hauser, "Generative AI enhances individual creativity but reduces the collective diversity of novel content", Science Advances, July 2024.

• PPC Land, "Self-ranked brands gain AI recommendations from 4% to 7%, Scrunch finds", September 2026 (report on Scrunch research).

• Google Search Central, "Our latest update to the quality rater guidelines: E-A-T gets an extra E for Experience", December 2022.

-----

How this was written: Topic, angle, convictions: mine. AI assists the writing. I make the call.

Sabrina Bulteau is the co-founder of PingPrime.ai, GEO Expert and specialist in Narrative Authority in AI Search. She helps brands, institutions and media become the reference AI engines trust, cite, and repeat, not just one option among many, by working both sides of the signal: on-site (structure, narratives, architecture) and off-site (earned media, platforms, co-citations, sector press). She previously co-founded Be Connect (acquired by iO Group) and Sench, is active within CEC Belgium, and brings 25 years of experience in digital growth and strategic positioning.