I started writing this blog in 2018, mostly because I had things to say about technology and business that didn’t fit into client deliverables. Seven years and a pandemic and an AI revolution later, it seems like a good moment to take stock. Not a predictions piece or a trends roundup. More of an honest look at what’s changed, what hasn’t, and what we keep getting wrong about technology transformation.
What Changed?
The obvious answer is AI, and the obvious answer is incomplete. AI is the most visible change, but the deeper shift is that the relationship between technology and business strategy stopped being metaphorical. A decade ago, when we talked about technology driving business outcomes, it was aspirational. Now it’s structural. Organizations that treat technology as a support function are measurably falling behind organizations that treat it as a core capability. My 2018 point about technology needing to drive rather than just enable has largely been settled because the market forced the issue.
Cloud went from novelty to utility to cost problem to managed capability, roughly in that order. The conversations I have about cloud now are completely different from 2018. Nobody’s asking whether to use cloud. They’re asking how to operate it well, how to manage costs, and how to make architectural decisions that don’t paint them into corners. That’s a sign of maturity, even if the path there involved some expensive lessons.
The tooling improved dramatically. CI/CD pipelines, developer platforms, API management, data engineering, observability… the infrastructure for building and operating software is genuinely better than it was seven years ago. Teams can do things today that would have required a dedicated platform team back then. It’s easy to take for granted.
What Didn’t Change?
The organizational challenges are remarkably persistent. We still see the frozen middle, jurisdictional debates about who owns cross-cutting capabilities, and a gap between what leadership says it wants and what the incentive structures actually reward. In all these areas, we still tend to confuse motion with progress. These were problems then and they’re problems now. The specifics change, cloud governance yesterday, AI governance today, but the underlying pattern is the same: organizations struggle to adapt their operating models as fast as their technology.
Technical debt still accumulates faster than organizations address it. Sadly, the creative plumbing metaphor hasn’t aged a day. Sure, said plumbing is more complex now, spanning cloud infrastructure, data pipelines, and AI systems, but the dynamic is identical. Teams take shortcuts under pressure, the shortcuts compound, and eventually someone has to pay the bill. Because the bill always comes.
People still get treated as an afterthought in technology-driven change. Despite everything we’ve learned about culture and psychological safety, most organizations still design the technology solution first and figure out the people part later. The Great Resignation should have been a wake-up call. Too often, the lesson faded as soon as the labor market shifted.
What We Keep Getting Wrong
We keep mistaking the technology for the transformation. Every wave, cloud, digital, AI, follows the same pattern. A new capability emerges. Organizations rush to adopt it. They invest heavily in the technology itself and underinvest in the organizational readiness to use it well. The early movers discover that the hard problems are governance, skills, operating model, and culture. The fast followers eventually learn the same lessons, having repeated most of the same mistakes.
We keep treating each wave as if it’s the first time. The AI adoption challenges of the past two years are eerily similar to the cloud adoption challenges of five years before that. The skills gap, governance vacuum, shadow usage, vendor hype, and executive pressure to act before understanding. All of it rhymes. Yet organizations approach each wave as a novel problem, rarely drawing on what they learned from the last one. Institutional memory for technology adoption is surprisingly short.
We keep underestimating how long real change takes. Every technology transition looks fast from the outside and slow from the inside. ChatGPT reached 100 million users in two months, but meaningful enterprise AI adoption is a multi-year journey that most organizations are still in the early stages of. Cloud migration was “done” years ago, except most organizations are still optimizing and many are still migrating. We are impatient by nature and the hype cycles feed that impatience. The organizations that do well are the ones comfortable with the long game.
The State Of The Craft
I’ve been a consultant in this space long enough to see several full cycles now, and the pattern that gives me the most hope is that the conversation has gotten more honest. In 2018, the dominant mode was aspiration. We’re going to be digital, agile, cloud-native, and it’s going to solve all problems. In 2025, the conversations I have with clients are more grounded. They’ve been through enough waves to be skeptical of silver bullets. They learned, often painfully, that the technology is the easier part and the organizational adaptation is the real work.
The craft of technology strategy has matured too. The best practitioners I encounter now operate with a blend of technical depth and organizational awareness that was rare a decade ago. They understand that their job includes the people, processes, politics, and yes, the systems. They’ve internalized that the real work happens in the seams between teams.
We’re never going to stop getting some things wrong. The next wave will come, and organizations will rush toward it with familiar enthusiasm and familiar blind spots. But the baseline keeps rising. We know more than we did. We’ve built better infrastructure, better practices, and a deeper appreciation for the messy human side of technology change. Exciting times!
