The buying shift in professional services
Professional services has always run on reputation. A new engagement usually started with a warm introduction, a name dropped in a board meeting, or a partner someone trusted — and then a confirming search to check the firm was real and credible. That confirming step is where the change is happening. A growing share of the people who buy consulting, accounting, advisory, audit and management services — CFOs, general counsel, founders, procurement leads, operations heads — now run that step inside an AI assistant rather than a results page. They ask in plain language: "Which firms are best for carve-out accounting in the UK mid-market?" or "Who should a venture-backed company use for R&D tax credits?" The model reads the web on their behalf and answers with a short, named list of firms. Often the buyer never sees ten blue links at all — they take the shortlist the AI hands them and start from there.
This is not a fringe behaviour any more. In G2's March 2026 study of B2B decision-makers, 51% said they now start their research with an AI chatbot more often than with Google, up from 29% a year earlier, with ChatGPT leading at 63%. For a category sold on trust and standing rather than a feed of product specs, that shift lands hard: the firms the AI is willing to name are the ones that get the call.
Why "best [consulting/accounting/advisory] firm for X" is the battleground
The queries that decide who gets shortlisted in this industry are remarkably specific: best advisory firm for [situation], top accounting firms for [sector or stage], who handles [niche need] for [type of company]. Each one is an implicit request for a ranked, named shortlist, and the model has to build that list from what it has already read. It cannot sit through your pitch. It cannot weigh your partners' track record directly. What it can do is notice which firms are named, described and discussed across the sources it treats as credible — industry media, directories, professional bodies, awards lists, roundups and editorial coverage — and surface the names that recur in the right context.
That is why AI visibility for Professional Services is an authority problem, not a website problem. A polished service page on your own domain tells the model what you claim about yourself; it carries little weight when the buyer's question is "who is actually best for this." It matters even more here because B2B services rarely have one decision-maker. Gartner's research on B2B buying describes buying groups of five to sixteen people, and finds buyers spend as little as 17% of total buying time with any one vendor. A consensus shortlist has to survive several skeptical people doing their own AI-assisted homework. If your firm isn't part of the material those people read, you are invisible at the precise moment the committee is drawing up its list — and nobody notices the omission. They just contact the firms the model named.