Good Transformer
Your clients are judging your firm on AI
Clients have stopped asking whether you use AI and started judging whether you use it well. A growing share are ready to move their work to a firm that does, and here is where to close the gap they can see.
Your clients have started judging your firm on how well it uses AI. A growing number say they are ready to move their work to a firm that uses it better.
A year ago a client might have asked whether you used AI on their work at all. That question has moved on. Now they assume you do, and they are forming a view on whether you are any good at it.
This is a different worry from the familiar one about whether your own AI spending pays off. The pressure here comes from outside the firm. Clients can see a gap, and some will take their business to a competitor who has closed it.
The gap clients can see
The clearest read on this comes from Thomson Reuters, whose 2026 Future of Professionals report surveyed more than 1,800 people across law, tax, audit, accountancy and related fields in 62 countries. Two of its findings sit awkwardly together. Some 78 per cent of corporate clients now rate AI-driven quality improvements as very important or essential, and only 6 per cent believe most of their providers actually deliver them.
That gap is starting to cost work. Within a year, about a third of those clients expect to reconsider which firms they use. A third of that group say more than a million dollars of annual work would be in play.
Applied to the United States alone, the report puts the total under active review at around 143 billion dollars. The figure is American and the market is not ours, but the behaviour behind it travels. A client who decides their adviser is behind on AI is now willing to shop.
Thomson Reuters sells AI tools to these professions, so it has an interest in sounding the alarm. Set that against the numbers, though, and the direction holds. Clients have moved from curiosity about AI to judgement about it, and judgement is what decides who gets the next instruction.
What clients actually notice
A client never sees the six hours your team saved. Whether that saving turned into anything is the internal question about productivity and value, and it matters, but it is yours, not theirs. What the client sees is narrower and sharper.
They see that the draft came back the next morning instead of the next week. They see that the analysis caught something last year's adviser missed. They see a fee that no longer squares with work a machine now does in minutes. Those are the things that travel by word of mouth, and they are what a competitor who uses AI well can put in front of the same client.
It shows up differently in each field. For an accountancy firm, it is management accounts and a plain-English commentary turned round in a day. For a corporate finance team, it is a first-pass analysis that comes back faster and more thorough than the manual version.
In legal work, it is a contract review that flags the odd clause the client's own team had missed. In recruitment, it is a shortlist with candidate summaries sharp enough to act on that afternoon. In an agency, it is a campaign report that reads like analysis rather than a data dump.
None of these needs the client to know a thing about the tool. They simply notice the work is better.
We already use AI, so why would a client think we are behind?
Because using AI and being consistently good at it are not the same thing. The same report found that 74 per cent of professionals use AI every week, yet 91 per cent say their organisation still falls short of what the technology could do for them. Weekly use is easy. Reliable quality is not.
In most firms the reason is uneven. A few people are genuinely good with these tools and the rest are not, so the client's experience depends on whose desk their file lands on. One matter comes back visibly better, the next looks like it always did, and the client draws their own conclusion about the firm.
There is a quieter signal underneath it. Roughly a third of professionals in the survey admit to using AI tools their firm has not approved, and that figure climbs where people feel the firm is moving too slowly.
When capable people reach for their own tools to get the work done, that tells you the firm is behind them. The fix is to bring good tools and training to them rather than police the ones they found.
There is a human cost in the same numbers. Two groups feel it first: juniors whose work is now measured against faster, AI-assisted colleagues, and staff told to use AI with no time to learn how. Closing the gap is as much about giving them proper help as it is about protecting the client relationship.
Where to start
Begin with the work a client sees rather than with a tool. There are three steps, and the order matters.
First, pick the single client-facing deliverable where a faster, sharper version would be most obvious. Get that one genuinely good with AI from end to end before spreading effort across everything. One visibly better output does more for a client relationship than a firm-wide rollout nobody feels.
Second, let the improvement show. If the work is quicker or better because you use AI well, put the result in front of the client. When they ask how you did it, have a clear and honest answer ready. Hiding the AI wins nothing and reads as evasion when they find out.
Third, close the gap inside the firm between the few who are fluent and everyone else, so a client's experience does not turn on who opens their file. That is mostly a matter of habits and training rather than tools, and it is the part most firms skip.
This is the work we do in AI Lessons for Leaders: taking the tasks a firm already does for clients and making them genuinely better with AI, without losing the judgement clients pay for. A single session on your firm's real client work is a sensible place to start. The sector overview for accountancy and advisory sets out where AI tends to help first.
Pick the deliverable your clients see most often, and ask an honest question about it. If a competitor down the road used AI well on this exact piece of work, could they do it faster, sharper or cheaper than you can today? Where the answer is yes, that is where your next lost client is, and where the work starts. If you want a second pair of eyes on that answer, book a discovery call.
Common questions
Are clients really moving firms over AI?
Some are preparing to. In the Thomson Reuters survey, about a third of corporate clients said they expect to reconsider their provider relationships within a year, much of it driven by whether the firm delivers AI-enabled quality. This is a shift in intention rather than a wave of switching so far, which is why it is worth acting on now.
What do clients actually notice about AI?
Clients do not notice your tools or your time savings. They notice turnaround, accuracy and price: work that comes back faster, catches more, and costs less to produce than it used to. Those are the visible signals a client uses to judge whether a firm is keeping up.
We already use AI. Why would a client still think we are behind?
Because quality is uneven. Weekly use is common, but consistent quality is rare, and a client judges you on their worst recent experience with the firm rather than your best. If results depend on which member of staff handles the work, the firm looks behind even when parts of it are ahead.
Where should a small firm start?
Start with one client-facing deliverable rather than a tool rollout. Make a single visible piece of work genuinely better with AI, then widen from there. A small firm can often move faster than a large one on this, which is an advantage worth using while it lasts.