Professional services do not sell an off-the-shelf product. Law firms, accountants, consultants, engineering firms, marketing agencies, HR service providers: the work consists of research, analysis, reporting, advice and the administration that comes with it. The hours largely go into work that repeats in form but not in content: every case is different, but the steps to get through it are fairly fixed. Gathering data, structuring it, drawing initial conclusions, drafting a concept, testing that concept against rules or precedents. Whoever wins in this sector usually doesn't win on the cheapest hourly rate but on a combination of speed, reliability of the outcome, and the ability to demonstrate that proper thought has gone into it. Which of these three weighs heaviest depends on the type of service, the size of the client, and the risks at stake.
For years, delivery time in this sector was mainly a matter of staffing: how many experienced people a firm could put on a case. That shifts as soon as AI partly takes over the research and structuring work. Drafting concepts, searching through documents, making initial categorizations: this is work of which part is now already done by AI, with an advisor approving or rejecting it with reasoning. The remaining part — the interpretation, the substantiation of advice, the conversation with the client about what an outcome means — remains human work, and that will remain the case for the foreseeable future.
The shift affects the comparison between firms in two ways. First, delivery time disappears as a distinguishing feature once preparatory work is largely automated; everyone who has the tooling delivers at roughly the same speed. The distinction then shifts to the quality of oversight: who checks what the AI has produced, and how thoroughly. Second, the cost structure changes. If part of the hours disappears from a case, what an hourly rate still covers changes, and with it what a quote is assessed on. Firms that still base their rate entirely on the old hourly structure lose on price against firms that have already redistributed their capacity — a question addressed separately in the analysis of price pressure and where it comes from.
The difference between firms that already capitalize on this shift and firms that do not rarely lies in ambition. It lies in whether the work has been broken down into tasks that can be assessed separately. A firm that treats a case as a single whole has a hard time inserting AI anywhere without revising the entire process. A firm that has already split the case into recognizable steps — gathering sources, summarizing, testing, concluding — can determine per step whether AI takes it over, does it partly with oversight, or whether the step remains human work. This breakdown is also where comparability between firms runs aground: two agencies that promise the same thing on paper can be organized completely differently internally, with consequences for pace and rate that only become visible upon detailed comparison.
This dynamic is not unique to professional services. Comparable shifts in what tips the scale can be recognized in the way AI is changing the competitive position in education and in the analysis of winning dimensions in retail. In every sector the pattern is the same: the work that is most repetitive and best defined shifts first, and the comparison between providers shifts along with it.
This shift says something about how work is organized, not about personnel decisions. Whether and how a firm adjusts its staffing to a changed hourly structure is a choice for the employer, with its own legal requirements on which this page makes no statement. What matters here is a different question: on which dimension does a firm claim to win, and can that claim be substantiated with evidence against what competitors actually deliver.
Whether AI in your firm already takes over the preparatory work, runs partly with oversight, or is still entirely human work, differs per task and per case type. That question — which work in your organization can genuinely be taken over by AI — is answered per task by the work scan from FTE TO AI, independent of what competitors claim to do. This is also relevant in a merger or acquisition, where two organizations with different divisions of labor are combined; see the approach for comparing positions after an acquisition for that.
Identify what you believe you win on: speed, thoroughness, price, specialization. With the free dimension check, you see which of those claims can be defended with evidence and which still rest on assumption. The full benchmark, with the evidence matrix per dimension and the comparison with your peer group, is under construction.