Marketers have ranked AI search and LLM answer engines as their number one distribution channel priority. They have also, at most organisations, done almost nothing to prepare their content for it. That contradiction is not a funding problem or a skills gap. It is an accountability gap, and it is getting wider every quarter.
The diagnosis comes from Emarketer, which describes the situation plainly as a «strategic paradox»: budget intent exists, tactical execution does not. Content operations for AI search remain immature. Structured data is inconsistently implemented. Citation monitoring across ChatGPT, Perplexity, Claude, Gemini, and Google’s AI Mode is, at most organisations, either ad hoc or absent entirely. Marketers say AI search matters most. Then they go back to writing content the same way they wrote it in 2019.
How organisations arrived here
The shift happened faster than most content teams could absorb. Google’s AI Overviews moved from limited test to default search experience within months. ChatGPT crossed 100 million weekly active users before most brand content strategies had even acknowledged LLMs as a distribution surface. Perplexity went from niche tool to cited source for product research across categories from software to financial services. The channel matured; the content operations did not follow.
Part of the lag is structural. Search engine optimisation has an established owner in most organisations, a clear reporting line, and decades of tooling. Generative engine optimisation has none of that. There is no dominant audit platform. There is no agreed metric set. There is no job title that owns it by default. So it falls between SEO, content strategy, and digital marketing, with each team assuming one of the others has picked it up.
The other part is a measurement problem. A brand can check its Google Search Console rankings daily. Tracking whether Claude cites your white paper when a buyer asks a category question requires a different process entirely, one that most organisations have not built. When something is hard to measure, it tends not to get prioritised in practice, regardless of what the annual strategy document says.
The strongest case for the status quo
There is a reasonable argument that early-mover investment in GEO is premature. The citation logic inside LLMs changes with each model update. What causes ChatGPT to surface a source today may be irrelevant when GPT-5 ships. Perplexity’s ranking signals differ from Gemini’s. Investing heavily in optimisation for systems whose behaviour is not yet stable or fully transparent carries real risk of wasted effort. Some content directors have made a deliberate choice to watch the space mature before committing resource.
That argument has merit if you are talking about deep technical optimisation. It has almost no merit when applied to the basics: auditing whether your organisation appears in AI-generated answers at all, understanding which content assets are being cited and which are invisible, and knowing whether a competitor is occupying the answer space you believe you own. Those are not bleeding-edge activities. They are table stakes for any channel you have named as your top priority.
The real issue is ownership, not ambition
Calling AI search the top distribution priority without assigning a named owner to the outcome is not a strategy. It is a preference. Preferences do not produce quarterly citation audits. They do not produce content briefs structured around the specific questions buyers ask LLMs. They do not produce anyone who can tell a CMO, on a given Tuesday, whether the brand is visible or invisible inside AI-generated answers for its core category.
Emarketer’s recommended fixes are specific: run citation audits across ChatGPT, Perplexity, Claude, Gemini, and AI Mode on a quarterly basis; assign a named GEO owner; rewrite content briefs so they lead with question-answer structure rather than keyword clusters. Each of those fixes is operational and assignable. None of them requires a budget line before the next financial year. What they do require is a decision about who is responsible.
That decision keeps not getting made because the consequences of failing to make it are currently invisible. When a brand drops out of Google’s first page, the ranking report shows it by Thursday. When a brand stops appearing in ChatGPT’s answer to «what’s the best [category] tool for mid-market companies», there is no alert, no dashboard, no meeting. The loss is real. Nobody sees it.
There is a useful diagnostic question to put to any leadership team that has named AI search as a priority: who is accountable if the brand becomes invisible in AI Overviews by Q3? If the answer involves more than one name, or requires a pause to figure out, the accountability gap is confirmed. That question is not rhetorical. It is the starting point for fixing the problem.
For content and marketing teams, the practical implication is narrow and immediate. Before the next planning cycle, run a manual citation audit across the five major LLM surfaces for your top ten category questions. Document what you find. Bring the results to whoever controls the content brief template and whoever owns the SEO relationship. Use those results to make the ownership question concrete. Abstractions about AI search priority will not move anything. A spreadsheet showing your brand is absent from twelve of the answers your buyers are getting will.
