Max: Sixty-five percent of Indonesian research users are now using AI search tools, and eighty-one percent say they use them specifically to understand and explore — not to click through to a website and evaluate for themselves.
Anita: That second figure is the one that should stop people. If users are accepting a synthesised answer as the evaluation, what exactly has changed structurally for a brand trying to be found?
Max: The intermediary step has been removed. The website used to do the persuading. A user would arrive, read, compare, and form a view. That step no longer reliably happens. The AI system forms the view on the user's behalf, and the brand either shaped that view through what it published, or it didn't.
Anita: So the question isn't whether your SEO is working. It's whether you even show up in the answer at all.
Max: And the mechanism that puts you there has almost nothing to do with paid placement. According to Maverick Indonesia, approximately ninety-nine percent of the links cited by AI systems come from non-paid media. Paid media, which remains a meaningful lever in traditional search, contributes less than one percent of influence in AI-generated answers.
Anita: That should reset a budget conversation for any organisation that has built its digital presence around bought placement. Where is influence actually coming from inside that non-paid base?
Max: Three source categories dominate. Journalistic media accounts for twenty-seven percent. Corporate blogs and owned content account for twenty-four percent. Aggregators and encyclopaedic sources account for seventeen percent. Read the top two together and the strategic conclusion is direct: in B2B, AI discovery is a function of news media and own media, in roughly equal measure.
Anita: So credibility you've earned through journalism and clarity you've published yourself — those are now the two assets doing the work. That's a different brief than most communications teams are operating to.
Max: It is, and it's produced a successor discipline to SEO. Generative Engine Optimization — GEO — is the practice of formatting content so that AI systems feature your brand inside generated answers rather than merely listing you among links. It rewards structure over keywords, factual precision over persuasion, and consistency over volume.
Anita: That inversion matters. Most content strategies are still optimised for human persuasion — narrative, tone, emotional framing. You're describing a system that rewards something closer to reference-grade clarity.
Max: Exactly. And the way to think about that practically is through what Maverick describes as the AI recommendation funnel — three distinct gates. Visibility: can AI find and mention your brand at all? Understanding: can AI accurately describe what you do and who you serve? Validation: can AI find evidence to support the claims you make?
Anita: Those three gates sound sequential, but I'd guess most organisations haven't formally tested any of them. Is there a measurement problem here as well?
Max: There is, and it sits behind all three. As AI answers become more prominent inside search, appearing in an answer doesn't reliably mean the appearance is recorded. A brand can be mentioned, surfaced, linked, or clicked, and each of those represents a different standard of proof. Visibility cannot be judged only by whether a brand appears — it has to be credited clearly enough to be seen, counted, and trusted.
Anita: So the measurement gap isn't just a technical problem. It's a signal that the content itself may not be structured clearly enough for the AI to confidently connect the right source to the right message.
Max: That's the right framing. The remedy is structural rather than promotional. Factual, consistent, well-organised content allows AI systems to make that connection reliably. And sitting behind all three funnel gates is a question most organisations have never formally answered: what is your customer actually likely to ask about you? Visibility against the wrong query is not visibility at all.
Anita: So the practical mandate isn't just publish more — it's map the query, then build content that passes all three gates against that query. What's the governing position here for a leadership team deciding where to focus?
Max: Our line is this: discovery has bifurcated from workflow, and leaders who treat AI purely as a productivity tool are missing the channel through which their market forms its view of them. With AI answers built almost entirely from earned and owned editorial rather than advertising, visibility, understanding and validation are no longer marketing metrics — they are corporate assets, and they behave like assets, compounding where they are maintained and eroding where they are not. For B2B organisations the mandate follows directly: you now market to humans and to machines simultaneously, and the single most important factor across every medium and channel is consistency in clarity and messaging. The question we would put to any business is not whether your brand appears in an AI answer, but whether the answer it gives is the one you would have given yourself.