Together, these three forces don’t create a five-year strategic horizon. They create urgency.

Where most firms actually are

Our research reveals only 21% of accounting firms have a formal AI policy in place. 46% offer no AI training to their team whatsoever.

The interest is real, but the frameworks aren’t. That gap between enthusiasm and readiness is exactly where most firms are stuck, and it’s where the competitive risk lives.

To understand how to close that gap, it helps to have a map. At Karbon, we think about firm transformation in four levels.

Level 1: Traditional

Manual workflows, time-based billing, everything runs through the partners. The firm is reactive. Clients come with problems, you solve them, you bill for time. Margins typically run 30–40%.

Level 2: Emerging

Cloud adoption is underway. Workflows are starting to standardize. Some areas may have moved to fixed-fee pricing. You’re building better habits, but AI isn’t central yet.

Level 3: Optimizing

AI starts reshaping the actual work. Advisory services lead the value proposition. Pricing reflects outcomes, not hours. The firm is becoming proactive.

Level 4: Leader

Human and AI working in true partnership. Autonomous systems handle routine work end-to-end. Margins at 70% or above. 80–90% recurring revenue. Pricing is outcome-based.

Most leaders I speak with think they’re further along than they are. Use this framework as a genuine diagnostic, rather than a box-ticking exercise. Where are you actually?

What the transformed firm looks like

Let’s take a closer look at Level 4, because it’s not theoretical. Some firms are already building it.

A transformed firm runs at 70%+ operating margins. A traditional firm runs at 30–40%. That’s a structural change in how value is created.

The firm generates 80–90% recurring revenue. This is a platform business, not a project shop. Clients are on ongoing advisory engagements. Revenue is predictable. The team knows exactly what they’re delivering and when.

AI is embedded at every stage of the client lifecycle. Not in one tool. Everywhere. How you attract clients, how you onboard them, how you deliver work, how you manage the relationship over years.

CAS firms used to sell clean books. Now the most forward-looking ones sell financial clarity. The firms that can deliver that consistently, at scale, through AI are the firms that will own the next decade of this profession.

Jobs are being transformed

The new model is AI and your team, each doing what they’re actually built for.

Agents are genuinely good at work that doesn’t require human judgment: collecting and classifying client data, executing recurring transactions, reconciling accounts, drafting communications, monitoring deadlines, chasing missing information. Your team is doing all of this right now. It doesn’t require their expertise. It requires consistency and scale.

Your people shine at the work that actually builds a great firm: applying professional judgment, advising clients on decisions that matter, leading relationships that go beyond the transactional, setting strategy, thinking about growth.

Before AI, your team was the doer AND the reviewer with no bandwidth left to be the advisor. With AI agents handling the process work, they become the reviewer AND the advisor. That’s a meaningful upgrade for everyone involved.

The firms that help their teams make this transition will attract and retain the best people. The firms that don’t will find themselves with a growing gap, not just in efficiency but in talent.

6 steps to start your AI transformation

So what do you actually do next? Here’s where I’d start.

1. Assess your level honestly

Map where your firm sits on the four-level framework. Most leaders discover they’re between Levels 1 and 2. This is a great starting point.

2. Build your AI policy

Only 21% of firms have one. Your team and your clients need to know there are guardrails in place. It gives you permission to move faster. Without it, you’re asking your people to make judgment calls they shouldn’t have to make alone.

3. Train your people to lead agents

This is the step firms consistently skip, and it’s why adoption stalls. Buying a tool and handing it to your team is not a transformation. Our research shows that firms investing in structured AI training save 28% more time than those who don’t, an additional 1.5 weeks of productive capacity per person, per year, from training alone.

4. Start with one workflow

Pick something repetitive, high-volume, and already somewhat standardized. Get it fully working. Measure the time saved. Use that win to build momentum and internal credibility for the next one.

5. Choose software partners built to last

A lot of AI vendors from 18 months ago are already gone. When you’re making decisions that affect your client data, your workflows, and your team’s daily experience, vendor stability matters. Prioritize purpose-built, financially sound partners over generic tools with AI bolted on.

6. Validate security and compliance

Your clients trust you with their most sensitive financial data. Every tool in your stack needs to earn that trust before you deploy it. This is not optional.

Transformation isn’t a project you launch. It’s a habit you build. Start with one step, get it right, and compound from there.

The window is open, but it’s narrowing

AI-enabled firms are already generating the equivalent of seven additional weeks of productive capacity per team member per year compared to firms that haven’t adopted AI. 63% of accounting professionals believe their firm’s value will decrease if it doesn’t adopt AI.

The firms that thrive in the next decade won’t just use AI. They’ll be built of it.

‘Using AI’ means adding tools at the edges. Being ‘built of AI’ means it’s structural, embedded in how you hire, how you price, how you serve clients, how you run the firm day-to-day.

Some of you reading this are already on that journey. For those who aren’t: the window is open, the roadmap exists, and the technology is here. The only variable is the decision.