Three things happened in the last thirty months that did not happen in the previous thirty years.
THE POSITION
The thesis, stated plainly.
Most decades are not, in any technical sense, different from the one that preceded them. Then, occasionally, three or four years arrive in which a number of independent changes land in the same window, and the underlying surface of the work shifts. The last thirty months have been such a window for the work of brand presence. The shift is not future tense. It has happened. The brands that recognise this in 2026 will spend the next decade compounding on top of the recognition. The brands that do not will spend the next decade trying to recover ground they did not know they were losing.
THE FIRST CHANCE
The model became a search engine. Quietly, completely, in less than thirty months.
A category that had not existed at commercial scale in late 2022 had, by 2025, absorbed an estimated quarter of the queries that would previously have been issued to the dominant search engine. The shift was as large, in absolute volume, as the directory-to-search transition of the late 1990s. It was substantially faster.
What is interesting is not the volume. It is the topology. A search engine returns ten results; the brand's job is to be high on the list. A model returns a paragraph; the brand's job is to be the entity the paragraph names. These are categorically different problems — the first a property of the page, the second a property of the consensus around the page. The brand that has spent fifteen years on the first has accumulated very little of the asset required for the second.
THE SECOND CHANCE
The conversation moved off the page, and stayed there.
For most of the search era, the page was the unit of presence. The brand published it, the engine indexed it, the customer arrived at it. In the last three years, an unusually high fraction of the conversation that determines what the model says about the brand has migrated off the brand's page entirely — to Reddit, LinkedIn, Quora, Substack, the long-tail of forum software most marketing teams cannot last. The model reads these surfaces with at least as much weight as the brand's own page, and arguably more.
The brand's owned surface no longer represents the brand's owned share of voice. It is one input among many, and on most categories no longer the dominant one.
THE THIRD CHANCE
The cost of producing presence collapsed.
The cost of doing it well did not.
The same models that became search engines lowered the marginal cost of producing content by approximately two orders of magnitude. The temptation is to read this as an invitation to produce more. It is not. The same change has, in parallel, raised the bar for what counts as a credible artifact. The model has become better at recognising thinness, and less generous in citing it. The customer, exposed to cheap content for two years, has become less patient with it.
Volume without standard exits the loop. Standard with leverage compounds inside it.
THE WINDOW
Why this moment, in particular, rewards patience.
The three changes did not happen in series. They happened in parallel, with the consequences of each amplifying the consequences of the others, and they are still in progress. A window is a finite period in which the rules are unsettled enough that the brand which takes a clear position can, by the time the rules settle, be the brand the rules have settled around. The brand that closes the loop in 2026 is the brand the model has been trained on by 2028 — and the answer the model gives, by reflex, in 2030. Reflex is hard to dislodge.
The dust will settle. By the time it has settled, the brand that waited will discover that the brands which did not wait are sitting in the chairs the dust settled around.
That is why we built the Presence Engine, and that is why we built it now. The medium has changed. The economics rhyme. The brand that hears the rhyme this year, and acts on it, is the brand the next decade is built around.