Why AI Efficiency Is a Race to the Bottom — and What You Should Be Building Instead

Every firm I talk to is using AI now. That part is settled. The question that isn’t settled — and that I think will define who wins the next decade in professional services — is what they’re actually using it for.

Most are using it to drive efficiency. To go faster. To do the same things they’ve always done, but with less friction. And I understand why: It’s the obvious first move, and it delivers real results in the short term.

But it’s not a strategy. It’s a race to the bottom disguised as progress.

The Blockbuster Problem

There’s a specific real-world example I’ve found that puts the problem into perspective: the fall of Blockbuster.

Blockbuster dominated the movie-rental industry. They were a household name, the go-to place for individuals and families to pick up in-home entertainment with ease. As times changed, they changed with them — to an extent. They adopted the technology that was in front of them, building a website and providing the latest physical movie formats. But they didn’t go beyond what was immediately attainable. They didn’t think to what changes were coming so that they could adapt the way they operated.

Their competitor, Netflix, did. First, with home DVD delivery, then with the instant-gratification of streaming. The rest is history.

Services firms are facing the same fork in the road right now. The firms that add AI to what they already do — making existing tasks more efficient, automating the routine — are Blockbuster with a better website. The firms that use AI to fundamentally change how they operate, how they package value, how they deliver and price their services? Those are the ones that will outperform.

It can’t be a race to the bottom. “My tasks have just become more efficient, but how am I growing my revenue? How am I expanding my market opportunity?” Those are the questions every leader needs to be asking.

What ‘Operating Model Change’ Actually Means

When I say firms need to transform how they operate with AI, I want to be specific about what that looks like, because it’s easy to nod at in a conference room and much harder to execute.

I spoke to one of our customers recently who now has their first AI-enabled, end-to-end service proposition and flow, which they’re now packaging and selling differently. That’s what the shift looks like. It’s not: we use AI to write status reports faster. Instead, it’s: we’ve rebuilt what we sell and how we sell it, and AI is foundational to the delivery model.

That requires a different kind of investment. Not just in tools, but in how you think about your business. What are the services you offer? Which can be productized and delivered at scale with a hybrid human-and-agent team? Where does the high-value human judgment live, and how are you protecting and compounding it?

Customers are already demanding this shift. I hear it in conversations every week: I’m not going to pay you for something I can use Claude or AI to do. Where’s your deep expertise and the value you’re selling to me? And then, how are you using AI to give me deeper value, quicker, faster, cheaper?

That’s not a future problem. That’s happening now.

It can’t be a race to the bottom. It has to be about how you change the way you operate as a business. To me, it comes back to that one thing that services firms and professional services really need to compound, which is your expertise.

The One Thing You Actually Need to Compound

There’s one asset that services firms have that AI simply cannot replicate: expertise. The accumulated knowledge of how to deliver, what works in a given industry or context, what the edge cases look like, how to spot when something is going wrong before the client does.

But here’s the thing: expertise only becomes a durable advantage if you treat it as something to compound, not just deploy. It can’t just become a commodity that’s automated with AI. It has to be something that you accumulate and build on. You’re continuously learning, and therefore continuously coming up with new propositions and new value.

That’s the real opportunity that most firms are missing. They’re using AI to do existing work faster. The firms that win will use AI to make their expertise smarter with every engagement, so that scoping gets better, delivery gets tighter, and new propositions emerge from patterns that would have been invisible before.

The Window is Open, but Not Indefinitely

I want to be direct about timing. There’s a tendency in industries facing disruption to say, “let’s wait for the technology to mature.” But the window is now. The organizations I talk to and work with are already moving, they’re shipping. The old model is already dying.

If you’re a consulting business that isn’t changing how you price, how you sell, how you operate with AI, you need to make those investments now. Because other organizations already are — and they’re going to outpace you to the point where you may not be able to catch up.

Efficiency gains bought you time. They didn’t buy you a future. The firms building something new with AI — new propositions, new delivery models, new ways of compounding expertise — are the ones that will still be winning five years from now.

The question isn’t whether to change. It’s whether you’re changing fast enough.