Short answer
In professional services firms, AI pays first where it moves the numbers the firm runs on: proposal win rate, hours per deliverable, time until new hires are billable, and non-billable reporting time. Thomson Reuters found organization-wide AI use in professional services nearly doubled to 40% in 2026, but only 18% of respondents knew their firm tracked the return. Pick one of those numbers, baseline it and check every release against it.
Key takeaways
- Organization-wide AI use in professional services rose from 22% to 40% between 2025 and 2026, per Thomson Reuters.
- Professional services economics connect AI directly to the income statement through hours, win rate and margin.
- The first places to look are proposals, research and first drafts, onboarding, client reporting, and being found in search and AI answers.
- A draft that's 90% right can cost a senior reviewer more time than writing it, so measure quality as well as speed.
AI use in professional services has moved from experiment to routine. Thomson Reuters reported in February 2026 that organization-wide AI use across the professional services firms it surveyed nearly doubled in a year, from 22% to 40%. The same survey found only 18% of respondents knew their organization was tracking the return. For consulting, accounting, engineering and marketing firms, the question now is where AI moves the numbers a firm actually runs on: utilization, realization, win rate and the hours senior people spend on work a system could draft.
Why professional services is different
A professional services firm sells expertise by the hour or by the project. That makes its economics unusually direct. If AI cuts the hours a deliverable takes, the firm either keeps the margin on fixed-fee work or has to rethink how it prices hourly work. If AI raises proposal win rate, revenue follows. Few industries can connect AI to the income statement this cleanly, which is why it's worth measuring properly. See how to measure the ROI of AI.
Where AI pays first
These are the places we'd look first, framed as the number each one moves.
- Proposals and pitches (revenue). Drafting from past proposals, case studies and the client's own material. The number is win rate or proposals completed per week.
- Research and first drafts (margin). Pulling sources, summarizing findings and producing a first version for a senior person to edit. The number is hours per deliverable.
- Onboarding and knowledge (hours). Answering "how do we do this here" from the firm's own documents. The number is time until a new hire is billable.
- Reporting to clients (margin). Turning project data into status reports and summaries. The number is non-billable hours per project.
- Being found (revenue). Content that shows up in Google and in AI answers when buyers research a problem. The number is booked calls.
What goes wrong
The common failure is buying a tool for everyone and measuring adoption. Seats get used, nobody can say what changed, and the renewal becomes a debate. The second failure is quality: a draft that's 90% right can cost a senior reviewer more time than writing it from scratch. Both are avoided the same way, by picking one number, measuring the baseline and checking every release against it. That's revenue-focused AI consulting in practice.
How Cosmic works with professional services firms
We implement AI to move one number, revenue, margin or hours, with a free 30-minute call first and a fixed-scope two- or four-hour discovery session with your executive sponsor. Each sprint has a fixed price plus an outcome-linked fee. If your own team will run the system afterwards, we pair with them until they can. We work with private firms across the US and in person across DC, Maryland and Virginia; we don't take government agency work.
For choosing a partner, see how to choose an AI consultancy, or fractional chief AI officer vs. an AI consultancy if what you need first is someone to own the firm's AI agenda.
FAQ
How are professional services firms using AI?
Mostly for drafting and research: proposals, first drafts of deliverables, summaries, internal knowledge search and client reporting. Thomson Reuters found organization-wide use reached 40% of the professional services organizations it surveyed in 2026.
Where should a consulting or accounting firm start with AI?
With the number that matters most this year, such as proposal win rate or hours per deliverable. Measure its baseline, put AI where it relieves that constraint, and report each release against the baseline.
Does Cosmic work with professional services firms?
Yes, with private firms across the US and in person across DC, Maryland and Virginia. Cosmic doesn't take government agency work.
Sources
Founder of Cosmic. Nearly 15 years in strategy, management and sales as a digital strategist, director of sales and product manager. He closed over $25M in new revenue at VMware Pivotal Labs, launched digital products at Capital One, and helped Viget grow from 30 to 75 people. He started Cosmic and built the Revenue Design® process after advising friends whose companies struggled to close deals and keep revenue steady.