I have personally witnessed every major technology adoption wave since the personal computer. The minicomputer. The Mac. The commercial internet. Streaming video. Smartphones. Cloud computing. Mobile-first. And now, AI.
In 40 years, the technology has changed every single time. The human behavior around adopting it has not changed once.
That pattern — consistent, predictable, and almost never discussed honestly in business circles — is the reason most AI implementations fail. And it's the reason a small number of companies will use AI to pull so far ahead of their competition in the next 24 months that the gap may never close.
I want to tell you exactly what that pattern looks like. Because I've lived it. Multiple times. From the inside.
The FedEx Scanner Nobody Wanted
Early in my career, I was a fast-tracked IT executive at Fderal Express — the era when Fred Smith was building something genuinely unprecedented in logistics, and the company was electric with innovation. I had the unusual privilege of interviewing senior executives including COO Jim Barksdale, who later became CEO of Netscape, as part of a major infrastructure assessment project.
At the same time, something was happening on the ground that I watched closely: the company was rolling out handheld scanning devices to delivery drivers. Every package, every stop, every handoff — scanned.
The reaction from drivers was immediate and fierce. "This is crap. It's going to slow us down. We don't need this." The resistance wasn't quiet grumbling. It was loud, sustained, and completely sincere.
A short year later, those same drivers would have shot anyone who tried to take those scanners away from them.
I have told that story dozens of times over 40 years. Every time I tell it, the person I'm telling it to nods — because they've seen the exact same thing with a different technology. The Mac. Email. The company website. HubSpot. Smartphones. Every single wave, the same arc: fierce resistance, grudging adoption, complete dependency.
The technology always wins. The question is never whether adoption will happen. The question is whether your organization is positioned to lead that adoption or scramble to catch up.
I've Watched This Movie Four Times
The FedEx story was just the opening scene. Here's the full film.
Wave 1 — The Internet. When the commercial web emerged in the mid-1990s, I was running a technology consulting firm and watching my clients split into two camps almost immediately. One group said "we need to be online" and figured it out, often imperfectly at first. The other group said "our customers don't use the internet" — and said it right up until they were watching their customers buy from competitors who did. The companies that survived that wave weren't the ones who moved first. They were the ones who moved with intention before the window closed.
Wave 2 — Mobile. I watched the same split happen again when smartphones became business tools. Companies that figured out mobile-first customer experience kept pace. Companies that treated mobile as a smaller version of their website fell behind in ways that compounded for years.
Wave 3 — Cloud and SaaS. The pattern repeated. The resistance was different — "security concerns," "we need to keep data on-premise" — but the outcome was the same. Early movers built operational advantages that late adopters spent years trying to close.
Wave 4 — AI. This is where we are right now. And I want to say something clearly: this wave is not different. The technology is more powerful than anything that came before it, yes. But the adoption pattern is identical. The same camps are forming. The same arguments are being made. The same gaps are opening.
The companies that will win with AI are not the ones moving fastest. They're the ones moving most honestly — about where they actually are, what's actually working, and what needs to change before they go further.
Why Most AI Implementations Are Failing Right Now
I've done enough technology implementation work — and enough post-mortems on implementations that didn't deliver — to see the failure modes clearly. They're not mysterious. They're almost always one of three things.
Failure Mode 1: Tools without strategy. A company buys an AI tool because a competitor has it, or because someone at a conference said they needed it, or because a vendor made a compelling demo. They implement it without mapping it to a specific business problem. Six months later, usage has dropped, the team is frustrated, and leadership concludes that "AI doesn't work for us." The problem was never the tool.
Failure Mode 2: Automation on top of broken processes. A company takes a dysfunctional workflow — unclear ownership, bad data, inconsistent execution — and automates it. AI doesn't fix broken processes. It accelerates them. Every inefficiency, every bad habit, every gap in the system moves faster. The implementation "fails" when in reality it simply exposed what was already there.
Failure Mode 3: Adoption without accountability. The tool gets purchased and deployed. Training happens once. Then leadership checks the usage dashboard three months later and discovers that 20% of the team is using it regularly, 80% reverted to what they were doing before. No one was held accountable for the change. No one measured what "success" actually looked like. The investment sits unused.
The companies succeeding with AI right now share one characteristic: they diagnosed their actual situation before they started. They knew what problem they were solving, what their current systems could and couldn't do, and what behavior change was required. They treated AI adoption the same way they'd treat any significant operational change — with rigor, with accountability, and with realistic expectations about timeline.
What This Means for Your Business Right Now
If you're reading this and nodding, you're probably in one of three situations.
You've already implemented AI tools and aren't seeing the results you expected. That's recoverable — but it requires an honest look at whether the problem is the tool, the strategy, or the adoption process.
You're in the planning phase and trying to figure out where to start. That's actually the best place to be. Starting with a clear diagnostic — what are we actually trying to accomplish, what do our systems look like today, what does our team need to be able to do — saves enormous time and money downstream.
You haven't started yet and you're feeling the pressure to move. That pressure is real, but speed without direction creates expensive messes. The companies that will fall furthest behind aren't the ones moving carefully. They're the ones moving fast in the wrong direction.
The technology is rarely the hard part. The hard part is people — getting humans to change how they work, overcoming the organizational friction that greets every new tool, and building accountability structures that make adoption stick.
That's not a comfortable thing for a technology company to say. But it's been true in every wave I've lived through, and it's true in this one.
A Practical First Step
The most common mistake I see companies make at the start of an AI initiative is skipping the diagnostic phase. They go straight to tools without understanding their own baseline — what's working, what's not, where the gaps are, and what AI could realistically change.
At 30dps, we call this a HubSpot AI Optimization Audit. It's not a sales conversation. It's a structured look at where you actually are: your current HubSpot setup, which AI features you're already paying for but not using, where your data quality is undermining your results, and what a realistic roadmap looks like for your specific situation.
If you're navigating these decisions right now — whether you're trying to fix a stalled implementation or trying to start one on the right foundation — that's exactly the conversation I'd like to have.
→ Learn about the HubSpot AI Optimization Audit