We've Made This Promise Before

AI is being sold as the technology that will finally remove the friction from product creation: faster concepts, fewer rounds, less waste, shorter calendars. The promise is everywhere, and much of it is believable. I believed it about Digital Product Creation, too. The technology was real. The promise was reasonable. Yet after a decade of investment, many organizations found themselves asking why the results hadn't matched the expectations. The pattern was surprisingly consistent: A new technology arrives promising to remove friction. It gets deployed before anyone defines the friction it's supposed to remove—and ends up creating a new kind of friction instead. The technology usually isn't the problem. The problem is that we skip the one step that determines whether the technology will ever deliver on its promise. Let me show you the same story twice. First with the wave we've already lived through. Then with the one arriving now.

The wave we've already lived through: DPC

Digital Product Creation was the move to conceive, develop, and decide on product digitally instead of physically. It didn't arrive all at once. It came in layers, over years—and the same mistake repeated at every one.

It started with 3D. The case was clean: cut fewer physical samples, see the product earlier, take rounds out of the calendar. All of it was true. Then the tools went into teams, and a friction no one had budgeted for showed up. Brands ran 3D and physical in parallel, because nobody fully trusted the screen yet. Instead of replacing a step, they added one. A render got treated like a specification, and decisions got made on an image that was never built to carry them.

None of that was a flaw in the software. It was the cost of deploying a capability before anyone defined the problem it was there to solve.

Then the promise got bigger, and the friction grew with it. A "digital twin" was supposed to be the product itself, expressed in data. But ask ten people in a footwear company what a digital twin is and you'll get ten answers. The vendor means a high-fidelity render. The designer means a sampling replacement. The manufacturing director means a build specification. The CFO means cost reduction. None of them wrong. None of them describing the same thing—and all of them funding the same line item against a different picture of what they were buying.

Then came the full program, and it had the same shape every time. Tools bought separately that don't talk to each other, so the single source of truth becomes three. Retraining no one scoped. Ownership of the digital asset left undefined, so the physical process never actually goes away. And because the friction was never named at the start, no one could say at the end whether it had been removed—so programs that delivered real value still read as failures.

The pattern didn't run once. It ran at every layer. A decade of chances to learn the shape of it.

Here it is again.

The wave arriving now: AI

AI is the next wave, and it's following the same pattern—only faster.

Organizations know they should be using AI. Far fewer can clearly articulate where it creates value inside their product creation process. That gap—between knowing you need a technology and understanding which problem it should solve—is where friction begins.

I've heard two versions of the same story repeatedly from people still inside the industry.

The first is organizational. Teams are told to "use AI," but no one defines what success actually looks like. The technology becomes a mandate instead of a solution.

"Use AI" lands as a mandate, not a problem to solve.

The second is familiar to anyone who lived through the early days of 3D. Business teams know where work feels slow but struggle to describe the root cause. Technology teams understand the tools but can't translate vague frustrations into targeted solutions. The AI ends up sitting between the two groups, satisfying neither.

Imagine a footwear design team using generative AI to produce dozens of concept sketches in minutes. Design celebrates the speed. Product management assumes the concepts are ready for development. Engineering quickly discovers they ignore manufacturing constraints, material limitations, or costing realities. Instead of eliminating work, the organization creates another review cycle to separate inspiration from execution.

The AI worked exactly as designed.

The organization hadn't decided which decision it was supposed to accelerate.

Neither of these failures belongs to the technology. They're consequences of deploying a capability before defining the friction it exists to remove.

That's the same mistake we made with 3D. It's the same mistake we made with digital twins. AI is simply repeating the pattern on a much faster clock.

Buying technology feels like progress.

Defining organizational problems feels like work.

Too often, organizations choose the former and postpone the latter.

Breaking the pattern

The pattern has only ever broken one way: someone defined the friction before they bought the tool.

That's the intervention.

Before deploying a 3D capability, a digital twin program, or an AI platform, answer one question:

What specific friction is this investment supposed to remove?

Everything else follows from that answer.

"We're investing in AI" isn't a strategy. It's a budget line.

The organizations that succeeded with Digital Product Creation weren't necessarily the ones with the most advanced software. They were the ones that identified the decision they wanted to make faster, aligned around it, and could measure whether the investment actually improved it.

This is fundamentally a design operations challenge—not a technology purchase.

Naming the friction. Defining success. Establishing ownership. Determining the new source of truth. Preparing the organization before the tool arrives.

That's the work that determines whether any technology wave delivers lasting value.

Every wave promises speed.

The organizations that benefit most aren't the ones that adopt first.

They're the ones that know exactly which decision they're trying to make faster.

Technology doesn't remove friction.

People do—by deciding which friction matters before the technology ever arrives.