Google DeepMind has demonstrated mathematically that one model can rank results without a separate index — a proof that, if it ships, ends the results page as marketers know it.

Search Engine Journal columnist Duane Forrester laid out the stakes bluntly: "Google DeepMind proved one model can theoretically rank without a separate index. If it ships, visibility stops being a position and becomes inclusion or absence."

That sentence compresses the entire argument. Ranking today presumes an index — a stored, crawlable corpus against which documents compete for positions. The DeepMind work, as reported by Search Engine Journal, shows a single model could perform ranking without that intermediate layer.

What changes if the index disappears?

The mechanics shift. A conventional pipeline crawls, indexes, and ranks. The DeepMind result suggests the ranking function could subsume the retrieval step entirely, at least in theory.

For search marketers, the consequence Forrester draws is categorical rather than incremental:

• Visibility stops being a position on a page.
• Visibility becomes binary: inclusion or absence.
• The results page, as a container of ranked links, does not survive the transition.

Why 'inclusion or absence' reframes SEO

For two decades, search optimization has optimized for relative placement — position three versus position seven, page one versus page two. Forrester's framing removes the gradient. A brand either exists in the model's answer space or it does not.

That is a different competitive problem. Placement tactics presuppose a page of results. If the model ranks without an index and answers without a ranked list, the addressable surface for optimization collapses to a single question: is the brand represented when the answer is generated, or omitted?

What remains theoretical — and what does not

The Search Engine Journal report is explicit about the conditional: the proof is theoretical, and its impact depends on whether Google ships it. "Theoretically" carries real weight in the original wording.

What is not theoretical is the direction. Google DeepMind, the company's AI research arm, has now published the mathematical groundwork. The gap between proof and product is an engineering and product decision, not a scientific one.

The so-what for brands

If Google deploys index-free ranking, the results page — ten blue links, then a ranked mosaic of widgets — becomes a legacy artifact. Marketing teams that measure success in rankings would need to measure it in presence: whether the model includes them at all.

Forrester's conclusion sets the agenda for the industry: watch whether Google converts this math into shipping infrastructure, because the day it does, "visibility stops being a position" — and every ranking report becomes obsolete.

Until Google confirms deployment, marketers face a proof, not a product. But the mathematics now exists, and Search Engine Journal's assessment is that the results page does not survive what comes next.