Unscripted SaaS · AI search
AI Search Is Not SEO 2.0
with Brittany Trafis — founder of Saurian Digital, an AI-native marketing agency
Host Jeremy Rivera gave Brittany Trafis a soapbox and asked for her hottest take — the biggest mistake business owners are committing that they should stop. She had one ready, and it was aimed squarely at the agencies serving them.
She runs Saurian Digital, which she describes as an AI-native marketing agency helping companies show up in generated results rather than only in traditional search. Before that she spent a decade helping lead a large digital marketing agency, which is the experience the take is built on.
The Frame That Is Costing SaaS Teams Visibility
“The number one piece I say is AI search is not SEO 2.0. And so SEO as a foundation is still important, like you outlined. Your backlinks, your keywords, that strategy still exists. And a lot of that is a foundation for AI search. So your technical components.”
— Brittany Trafis, Saurian DigitalNote what she does not say. This is not a claim that SEO is dead or that the fundamentals stopped mattering — she explicitly calls them a foundation. The claim is narrower and sharper: it is a different channel, and treating it as a version upgrade of the previous one produces the previous one’s results.
The soapbox version is blunter:
“My hottest thing that I’m hitting right now is quit treating AI search as SEO. Stop, stop doing it. If you keep treating it like just keywords and rankings, you’re going to get keywords and rankings. You will get keywords and rankings in Google. If you’re trying to show up in ChatGPT and Claude and Perplexity — where people are more engaged and want that information — you’re not going to see it. And so you have to stop treating it as the same channel.”
— Brittany Trafis, Saurian Digital“You will get keywords and rankings in Google” is the line to sit with. The work is not wasted; it simply pays out in the currency you already had. Teams doing this can show improving reports for a year while the new surface goes unaddressed, because nothing is visibly failing.
Ask Your Agency Three Questions Deeper
Her practical test is specific, and it is aimed at a claim almost every agency now makes:
“And if your agency, if your SEO agency says, we got it, we’re doing AEO, we’re doing AI search — really lean in, ask like three more questions deeper. And I think you’ll find out that they’re just treating it like SEO.”
— Brittany Trafis, Saurian Digital“Three more questions deeper” is a good heuristic because the first answer is always fine. Everyone can say the words. The distinction shows up at the second and third layer, where a keyword-shaped process has to describe itself in detail.
The three questions below are ours, built to apply her test. She names the test, not the questions.
“What are you measuring, and against which models?” An answer about rankings and positions is the keyword process wearing a new label. An answer that distinguishes between presence, accuracy and citation across several named assistants is not.
“What do you do when we are absent from an answer?” “Create more content targeting that query” is the old process. A diagnostic answer — which model, what it found instead, what source it used — is the new one.
“How often does this run?” A quarterly audit treats the surface as stable. It is not.
What Actually Moves Assistant Visibility
Brittany’s most concrete technical point is about schema, and specifically about the difference between having it and it doing work:
“You have to set up your schema correctly. But what the LLM cares about is what is that schema telling me? Like you think about a FAQ schema — AIs, they love it. They love it. It’s like, you’re going to tell me the question and the answer, and I can just grab that from your schema and I don’t have to go any further. Great. Because they’re trying to do it so quickly compared to where Google’s doing more of the rankings.”
— Brittany Trafis, Saurian DigitalThe mechanism she describes is about speed and cost, not preference: “The LLMs, they’re looking for the right response to the question that they have, and they’re moving very quickly. And so that’s why that technical element is so important.”
Which produces an unusually harsh failure condition:
“Now, if your schema is not set up correctly and it’s timing out, then it’s gonna move on and it’s going to ignore you.”
— Brittany Trafis, Saurian DigitalTiming out is not a ranking penalty you can diagnose in a report. It is silent exclusion. A system optimising for speed does not wait for your server and does not tell you it left.
She is also honest about the state of the field: “It is the wild wild west right now. And we are watching this channel be born… we’re still trying to settle on a name. You still hear AEO, GEO, AIO, AI search. I call it AI search as a whole.” A channel that does not yet have an agreed name is a channel where confident methodology claims deserve scrutiny.
Rebuilding The SaaS Content Plan Around It
Practitioner guidance, extending her argument.
If the channel is genuinely different, the plan has to change at the unit level, not the volume level.
Plan around questions, not keywords. The unit of an assistant interaction is a question with context, frequently a capability question — will it do this specific thing for me — rather than a category query.
Make answers extractable. Brittany’s FAQ schema point generalises: the easier it is to lift a clean question-and-answer pair from your page, the less work a model has to do, and it is optimising for exactly that.
Fix the technical floor first. Schema that times out is worse than no schema, because you have paid for it and are still invisible.
Keep the SEO foundation. She is unambiguous that backlinks, keywords and technical work still matter. This is an addition to the plan, not a replacement for it — and any agency selling it as a replacement has failed her test from the other direction.
Her own agency thesis follows the same logic. Rather than bolting AI onto an existing structure, she asked “what’s this look like if I build AI first and then people next? There’s still room for people. We still need the expertise. But it completely changes” the shape of the thing.
Full recap: Brittany Trafis on why AI search is not SEO 2.0. Saurian Digital is at sauriandigital.com.
For a compatible argument about the framing itself, see Stephan Bajayo on why “ranking in an LLM” is the wrong frame, and the measurement side in Somya Goyal on AI brand visibility.