AI-driven compliance will soon create a billing paradox in accountancy firms

Growth in accounting firms has traditionally followed a fairly reliable playbook: improve delivery, add capacity, hire good people, and build additional services around a solid compliance core. Most firms know the pattern well, and for a long time, it’s worked. Technology has helped too, but mainly by making existing work more efficient rather than changing what firms can actually deliver to clients.

What now seems to be shifting, gradually, is not just how work gets done, but where the real growth constraints sit inside the firm. It’s early, but the change is becoming noticeable.

It’s a sentiment I hear directly from partners within firms, who often say, “we’re getting more efficient, but it’s not making growth easier.”

I wouldn’t treat that as universal, but it comes up often enough to be worth attention. Progress is real, in that compliance delivery is faster, automation is reducing manual effort, data flows are cleaner, and turnaround times are improving. By most measures, firms are running tighter processes than before.

What’s more interesting is what happens after those efficiency gains arrive. In a number of firms, once compliance becomes more streamlined, another constraint stands out, which is that revenue still tracks closely to expert time and individual attention. Delivery improves, but scale doesn’t automatically follow.

Optimisation itself isn’t trivial. Aligning systems, redesigning workflows, and getting teams comfortable with new tooling takes real effort. But once operations stabilise, a broader leadership question tends to surface. If delivery becomes more efficient, where does scalable growth come from?

Advisory is “everywhere in theory, but accidental in practice”

In many of those conversations, “advisory” is where attention turns. It’s widely seen as the growth layer, where deeper client value and differentiation sit, yet also one of the hardest capabilities to extend consistently.

But it’s important to note that “advisory” doesn’t mean the same thing to every firm. Some use it for specialist, project-based work like transactions, restructurings, succession planning, major risk or strategy decisions, which are typically handled by senior experts. Meanwhile, others mean something more everyday, like practical guidance around pay structures, financing choices, hiring, margins, or working capital. That second layer appears frequently inside compliance interactions, but is rarely labelled or tracked as advisory.

The constraint is often structural rather than commercial. Advisory tends to exist unevenly, depending on timing, context, or a senior person spotting a signal. As one managing partner put it to me, it’s “everywhere in theory, but accidental in practice.”

A lot of everyday advisory is already embedded in routine client work, such as questions on salary mix, financing, hiring timing, margins, or working capital. Clients are asking for it but just not in a structured way. The insight exists, but it often lives in emails, calls, and personal notes rather than shared systems. When advisory depends on individual awareness and memory, it’s naturally difficult to deliver consistently across teams or offices. That’s not a criticism; it’s just how most firms have grown.

When efficiency reveals structural limits

AI is where this starts to shift, although it’s still early. Most AI discussions in accounting begin with time savings, but what’s more revealing is where broader signal visibility changes what firms can offer, not just how quickly they deliver it.

Across firms experimenting more deeply, a pattern is emerging: when client signals become more visible across emails, documents, financial movements, and conversations, advisory becomes less dependent on individual memory and more of a shared capability. Junior team members spot more, specialists engage earlier, and cross-service opportunities surface more consistently.

Not every firm is seeing this yet, but where it happens, the shift is noticeable. Advisory starts to become less episodic and more continuous, which means it’s less personality-driven and more infrastructure-supported, which may matter more than the raw time savings. It also changes how capacity feels inside the firm, because better signal detection doesn’t always require more client conversations, it just needs better visibility into the ones already happening.

A billing paradox

A billing paradox is starting to appear. If capability expands while pricing remains anchored to hours, the value equation can begin to feel slightly misaligned. Many firms describe this as a “good problem,” but still a real one, in that more expert capacity exists, yet the commercial model doesn’t always flex with it.

In response, some firms are beginning to experiment – though not uniformly and not without friction – with alternative structures. That includes more subscription-style relationships, value-based pricing for certain advisory outcomes, or packaging additional advisory and insight services alongside core compliance. Others are layering recurring advisory programmes or productised service bundles on top of traditional delivery.

It’s still early, but the common thread is that monetising expanded capability seems to depend less on the technology itself and more on how services are framed, packaged, and governed commercially.

A similar pattern shows up with AI pilots. Many succeed locally – tools perform well and teams save time – yet the wider organisational impact varies. What seems to separate the stronger outcomes isn’t model quality so much as whether surrounding workflows and decision rights evolve with the technology.

There’s unlikely to be a single right approach here, because firms differ too much in structure, client mix, and ambition for that. But the direction of travel does seem to be shifting. If advisory is increasingly treated as the growth engine, the practical challenge becomes how reliably it can be delivered across the firm, not just by individual experts at isolated moments.

AI won’t solve that by itself, but it is making both the constraint and the opportunity easier to see, which is often where meaningful change starts.