How AI search works for advisor firms

Short answers first, the evidence underneath, and a few things to try on your own firm.

Pieces that discuss the SEC Marketing Rule are marked.

How does the AI decide who to recommend?

With search on, it can search, read a few pages and answer from them.

Is there a page on your site that answers the question?

Consensus trust: Our term for the pattern in which AI assistants lean on what several independent sources say in the same way. A firm that many unrelated pages describe consistently gives an assistant more to agree on.
Citation: A link an assistant shows as the source for part of its answer. Being cited means your page was shown as a source; being named means your firm appears in the answer itself.
Fan-out: The separate searches an assistant can run to answer one question. Google uses the term for the related searches its AI features may run across subtopics.
Answer-first page: A page whose first two or three sentences answer one specific question, before any detail. It is a shape that is easy for an assistant to quote.
Entity description: The short account of who a firm is, what it does and where, that assistants repeat about it. When it is out of date, assistants can repeat the old version with confidence.

Jonathan Michael on AI inside the advisory practice.

Podcasts and press, with the real publisher named.