AI search has fundamentally changed SEO. Instead of typing keywords into Google and clicking blue links, people now ask ChatGPT, Perplexity, Google AI Mode, and Gemini full questions — and get one synthesized answer. The old goal of “rank #1 and earn the click” is being replaced by a new one: become the brand the AI mentions and cites.
To understand the shift on the ground, OtterlyAI CEO and co-founder Thomas Peham traveled across Germany — meeting marketing leaders, SEO experts, and builders in Hamburg and beyond – to ask one question: has anyone actually figured out how to show up as a brand inside AI search? This is what they said. (Watch the full documentary below.)
“At the beginning, SEO was basically our religion. And by now it’s probably AI search.” — Marvin Müller, VP Marketing at OMR Reviews
From Keyword Stuffing to AI Answers: A 15-Year Shift
Thomas Peham has watched this evolution from the inside. “In 2008, I wrote my bachelor thesis on SEO,” he recalls. “Back then, search engine optimization was a pretty wild, nerdy frontier. You could stuff keywords into a page and rank on Google.” Over the next 15 years — including building the marketing team at Storyblok, one of Europe’s fastest-growing CMS platforms — he watched SEO grow “from a small, nerdy thing into a real discipline.”
Now that discipline is being rewritten again. “This is the story of how search is being rewritten,” Peham says, “why most brands aren’t ready, and what I’m building to fix it.”
What Is Generative Engine Optimization (GEO)?
Generative Engine Optimization (GEO) is the practice of optimizing content so a brand gets mentioned and cited inside answers generated by AI engines like ChatGPT, Perplexity, Google AI Mode, and Gemini. Unlike traditional SEO, which aims to rank a page on a results page, GEO targets inclusion in the AI’s synthesized answer itself.
A common misconception is that the two are the same. As one SEO expert in the film puts it:
“Everyone says GEO is just SEO. I’d say SEO is the foundation of good GEO visibility – but ChatGPT and Perplexity work differently. It’s worth separating Google, AI Mode and AI Overviews from the AI engines.” – Matthäus Michalik, CEO at Claneo
Another leader is blunter about why copy-pasting your SEO playbook fails:
“Most of our customers are just trying to do the same things they do in SEO, for GEO. From my point of view, that does not work. It’s a completely different mechanism.” – TImo Mayer, Executive Board Member at Virtual Identity
Old SEO vs. AI Search: What Actually Changed
| Factor | Old SEO (Google links) | New AI Search (GEO) |
|---|---|---|
| Where users go | Search results page | ChatGPT, Perplexity, Gemini, AI Mode |
| The goal | Rank #1, earn the click | Get mentioned and cited in the answer |
| Success metric | Rankings, clicks, traffic | Mentions, citations, share of voice, sentiment |
| Authority signal | Backlinks, domain authority | Cited expertise, trustworthy content, entity clarity |
| Content shape | Structured for crawlers | Structured for LLM comprehension |
The biggest practical change: buyer behavior moved before the metrics did. “People do Google less,” Peham notes. “They start their journey in ChatGPT, they start their journey in Claude.” And the data backs it up:
“There’s data from the US that already 51% of software buyers start their research in AI.” — Marvin Müller, VP Marketing at OMR Reviews
Why Clicks Are No Longer Enough
For two decades, SEO measurement orbited a single number: organic traffic. AI search breaks that model, because the most valuable interaction now happens inside the AI answer, where traditional analytics can’t see. As one expert warns, dismissing AI search because referral traffic looks small is the central mistake:
“AI search is a brand channel — it’s not made to drive traffic. The biggest mistake is thinking ‘AI search isn’t relevant to me because I only have 2% AI traffic.’ Track the right prompts, track mentions, track citations, and track sentiment.” — Matthäus Michalik, CEO at Claneo
That reframe matters because the change is already showing up in pipelines. One marketing leader reported that demos attributed to AI engines jumped “from below 1% a few months ago to right around 5 to 10% last month.” The signal is real even when the raw traffic percentage stays low.
So the questions marketers need to answer are new:
- Mentions — How often does an AI engine name your brand when asked about your category?
- Citations — When the AI links to a source, is it citing you or a competitor?
- Sentiment — When your brand comes up, is the AI describing it positively, neutrally, or negatively?
- Share of voice — Across the prompts your buyers ask, how often do you appear?
- Prompts — Which questions surface your brand, and which leave you out?
The New Playbook: Trustworthy Content for Two Audiences
If GEO is a different mechanism, what actually works? The leaders in the film converge on a few themes.
Write for shoppers and for AI. As Tamara Sanderson frames it: “There are two customers now – shoppers and AI. We have to speak to both.” AI engines, she adds, are “prioritizing authentic, trustworthy content, and there’s no better trustworthy content than actual customer feedback.”
Keep the SEO fundamentals — then add GEO layers. Eugen Knippel described their stack as “a mix of traditional SEO thinking — H1s, H2s, the technical stuff — but also FAQs, lots of schema markup,” plus LinkedIn articles as “a credible and often cited source,” and PR agencies “to get high-value citations from other sources.”
Play more channels, because they compound. “The backbone is just having good content all over the internet,” Richard Wagentristl from Chatarmin says. “The more channels we play — even social like Instagram and Facebook — they all compound together.”
A welcome side effect: “I’m really happy that PR is back to the party.”
How to Start Optimizing for AI Search
- Map your prompts. List the real questions buyers ask AI about your category — not just keywords they’d type into Google.
- Measure your baseline. Track how often AI engines mention and cite your brand for those prompts, and the sentiment attached.
- Create citable content. Clear definitions, FAQs, comparison tables, schema markup, and source-backed claims are what AI engines quote.
- Build off-site authority. Earn citations from PR, LinkedIn, and review platforms — the trustworthy sources AI engines lean on.
- Monitor and iterate. Treat mentions, citations, sentiment, and share of voice like rankings — review them regularly.
Watch the Full Story
OtterlyAI is where marketing teams go to track and improve how their brand shows up in AI search – ChatGPT, Perplexity, AI Overviews, Gemini, and more. As Thomas Peham puts it: “Everything we learn, we publish.”
This is the story of SEO, and how AI has changed it.
▶ Watch the full documentary now — then see where your brand stands at otterly.ai.
