It struck us while listening to Mark Leonard that, for all the noise around artificial intelligence, some of the most experienced and thoughtful voices in the industry remain cautious. Leonard hardly mentioned AI at all — and yet, in that silence, there’s something worth reflecting on.
At Dellecod Assets Limited, we often look to companies like Constellation Software for signals about what actually matters in tech — not just what’s trending. And from their recent communication, there’s a subtle but important shift worth paying attention to: AI is being used, but not glamorized. It’s being integrated, but not idolized. That feels honest.
One of the more grounded observations coming out of Constellation’s work is that AI — so far — hasn't transformed industries. Most advances have been iterative. Efficiency is improving in areas like billing, customer service, and marketing. Better response rates. Smoother hand-offs. Tighter operations. But not reinvention. No vertical has been upended. No category rewritten by AI alone.
That’s not to say it can’t happen. But it reminds us of a critical distinction: AI is a tool, not a business model. Without deep context — the kind of operational and sector-specific knowledge that Mark Leonard’s organization has developed over decades — even the best algorithms don’t lead to much more than marginal gains. The magic, if we can call it that, lies in knowing your industry well enough to redesign how work gets done.
This is where we see the longer game. The opportunity isn’t in slapping AI onto existing systems and expecting disruption. It’s in redesigning workflows entirely — not just optimizing them — because you understand the jobs to be done at a granular level. And to do that, you need domain knowledge. You need data that’s been cleaned, vetted, and contextualized properly. Things that don’t make headlines, but quietly build moats.
Leonard’s restraint, then, may reflect not skepticism, but clarity. The foundational advantages aren’t in the tech itself, but in the conditions that allow it to matter. In an era obsessed with acceleration, there’s something almost comforting about taking the long view. We can admire the engineers and the breakthroughs — but we should also study the quietly persistent operators who make things stick.
At Dellecod, we’ve come to believe that successful deployment of AI begins with humility. It starts with the question: What do we actually know about this vertical? Not in theory, but in practice. What repeated frictions can we resolve, not just automate? And what parts of our workflow are so fixed that we’re blind to redesigning them entirely?
We don’t have all the answers yet. But it’s encouraging to see organizations like Constellation experimenting behind the scenes — quietly improving, patiently learning, making small dents. In an environment where disruption is often promised, it’s the quiet evolutions that may end up mattering more.
They build momentum slowly. They leave fewer headlines. But they tend to endure.