Technology Adoption Is Not The Same Thing As Transformation

Fashion has now spent a decade or more talking about digital transformation. Yet despite billions being invested in technology, countless pilot projects being run, and an endless stream of new platforms and ideas being introduced, many of the industry’s fundamental challenges remain unchanged. 

Development cycles are still too long. Products still miss their demand window. Teams still operate in silos, and reconcile their work when they get together afterwards. Sampling remains expensive, wasteful, and omnipresent. Decision-making is still fragmented – perhaps more so than ever. And every few years, a new technology arrives promising to solve it all.

First it was product data management, and later product lifecycle management: PLM. Then it was 3D. Now it is AI. These are all different proposals with different ambitions, and they have made different degrees of progress towards realising their visions (none have fully accomplished them), but the pattern that cuts across them  is familiar.

In my experience, the problem has rarely been the technology itself. The problem is that fashion has consistently approached digital transformation as a technology implementation exercise rather than a business transformation exercise, and the groundwork that’s missing from one project doesn’t then get laid for the next. Everything starts fresh, every time.

Too often, companies become fixated on the potential benefits of a new tool without fully understanding what would actually be required to unlock those benefits. They hear promises of faster development, fewer samples, improved efficiency, better decision-making, or lower costs. Someone shortlists, selects, and buys the solution, starts a pilot project, and everyone’s expectations are set sky-high… before the results roll in and fall short of the original promise.

The root cause here is seldom the technology itself being missold, user-hostile, or not fit for purpose capability-wise. That does happen, but it’s the exception rather than the rule. The delta between what everyone wanted to happen and what actually happened stems from the fact that the operating model remained largely untouched. The company now has another part of its technology toolkit, but its processes, decision-making structures, and operational setup haven’t been touched. The technology has simply been layered on top of the strata of systems, people, and silos that already existed.

Rarely does the industry stop to ask a much harder question: What if the technology wasn’t the problem? What if the real issue was that we expected a new tool to transform a business model that was never designed to support it in the first place? 

In practice, digital transformation has relatively little to do with the tools themselves. Technology changes constantly, and at least some of the solutions you’re implementing today are probably transient, and destined to earn a “legacy” designation in the long run. The companies that successfully navigate technological change are rarely the ones who prioritise  access to the newest tools above all else. They are the ones that build operating models capable of adapting as those tools evolve, cycle in and out, and even as we experience step-changes like AI has the potential to be.

Fashion, unfortunately, has often done the opposite. We have repeatedly tried to force new technologies into old structures, putting the tool in the driving seat, shaping the project around its demands rather than ours. Then we wonder why the promised transformation never arrived.

I believe this is what happened with 3D, which still has so much runway left unexplored, but which a lot of companies are now gearing up to try and discard in favour of something new. And if we are not careful, it is exactly what will happen again with AI. What seems novel and exciting today will turn into tomorrow’s letdown unless we treat it as a chance to do things differently.

What Fashion Can Learn From The 3D Adoption Journey

For years, 3D was positioned as the technology that would transform fashion. Companies invested heavily in software, training programmes, digital samples, asset libraries, and specialist teams. Conferences were filled with presentations about faster development, fewer samples, lower costs, and improved sustainability, and while those things were realised in some cases, the top-level vision for an all-encompassing digital twin still hasn’t materialised.

Looking back, I don’t think this gap persists because of 3D. I think the industry misunderstood what 3D actually was, and how and where to use it. I worry I see the same thing happening with AI, because the first steps along the journey already feel similar. 

From the very beginning, fashion largely positioned 3D as a design tool rather than something that would change overall operations. That may sound like a subtle distinction, but I believe it fundamentally shaped how the technology was deployed. The assumption was that designers and engineers would ideate in 3D, develop in 3D, iterate in 3D, approve in 3D, and eventually replace significant parts of the traditional development process through 3D, and there the value would stop. Later, “scaling 3D” became a synonym for attempting to apply the same discrete-discipline logic to marketing and communications.

The problem is that most commercial fashion businesses were never designed to operate that way, where a modality of working has a fixed start and end point, and where you can transfer so much responsibility from across the extended lifecycle to a small team that operates only at the beginning.

Speed and scale are the defining characteristics of commercial fashion, and meeting those goals means a lot of role-blurring, collaboration, tension, reconciliation and all-round messiness. No one team is responsible for hitting big strategic targets. 

Designers, too, are already expected to generate a huge number of ideas, react quickly to changing trends, and move products through development at an extraordinary pace. In that environment, asking designers to build every idea in 3D was always going to create friction, because it put a huge weight at one end of the product lifecycle, and because it didn’t then extend the tools to the rest of the organisation to help pick up any of that burden, or even to see how it was relevant to them at all.

What’s interesting is that this is immediately obvious when we look outside fashion. A lone architect doesn’t often take every single idea and build it into a fully developed 3D model. Hundreds of concepts may be explored through sketches, discussions, references, and quick visualizations before one direction is selected for deeper development. Only then does detailed modelling begin, and that work doesn’t typically fall on a single artist. 

The purpose of this digital-first model is not to generate every idea. The purpose of the model is to validate, communicate, refine, and approve the idea that has already emerged through an iterative, multi–stakeholder process that ran more smoothly because the various people involved had a common reference frame for making choices and recommendations.

Fashion somehow inverted this logic. Instead of treating 3D as a powerful approval and validation tool, the industry often tried to position it as the primary ideation environment, and then attempted to stretch it into downstream marketing afterwards. In doing so, many businesses ended up forcing the technology into workflows it was never naturally suited for – or at least into applications where the responsible teams were under-resourced and where the wider business was only tangentially aware of how their work contributed to the brand’s strategic goals.

Ironically, the outcome was that the very benefits companies were chasing (speed, efficiency, scalability, reduced physical dependency) became harder to achieve rather than easier, and that vision one source of truth (the digital twin) went unrealised.

And yet those benefits remain entirely real. They are on the table for the companies that want them, and that are willing to properly engage with what 3D is for and how it should be implemented – as part of a whole-business change initiative.

In fact, I would argue that many of the common “North Star” aims that fashion has today  can only be fully achieved through 3D – not  as a place where any idea the company entertains can come to life, but as the right place, and the correct container, for validating, approving, and then managing the right ideas through their entire extended lifecycles. 

AI Is About To Repeat the Same Mistakes

This is not just a history lesson in 3D. I’m a firm believer in the long-term potential of digital product creation workflows, but I am also acutely aware that the trajectory DPC found itself on a decade ago is one that AI finds itself at the start of today. 

It’s very clear that the technology is advancing at extraordinary speed. New tools emerge almost weekly andcompanies are racing to implement AI strategies to take advantage of those capabilities. Every conference agenda, industry report, and boardroom discussion seems to contain some version of the same question: “How can AI make us faster, cheaper, and more efficient?”

Those are reasonable questions. The problem is that they are often being approached in exactly the same way the industry approached 3D: from the tool first.

Instead of asking how AI could influence a positive, profound change to their operating model, many companies are asking how AI can be inserted into their existing ways of working. 

Across the industry, I see AI being layered on top of existing processes, existing approval structures, existing development workflows, and existing organisational silos. Does this sound like something we’ve heard before?

A disconnected process does not become connected because AI is added to it, any more than it did when 3D was the lens the industry was looking at itself through. A slow decision-making structure does not become agile because someone decided AI could supercharge parts of it. A fragmented, unsustainable business model does not become integrated and low-impact just because AI is added to it.

If anything, the opposite can happen.

When AI is introduced without a broader operational vision, it risks becoming yet another disconnected layer, and creating yet another source of information that exists on a tangent, independently of the rest of the business.

The result is often more complexity rather than less. More outputs rather than better decisions. More content rather than more clarity. More activity rather than more value.

There is also a line that AI could be about to cross that, to my mind, makes this early stage more contentious and more critical than it has been with other technology initiatives: the line between AI as an addition to the technology ecosystem, and AI as a replacement for 3D, design tool, development systems, decision-making platforms and so on. 

In reality, most of the meaningful value I see emerging from AI has very little to do with replacement. It has everything to do with amplification. 

AI can increase speed and adaptability. It can explode the number of options that can be explored before decisions are made. It can amplify human capabilities in a range of different areas. These are real, and potentially measurable outcomes, but they only create value when they exist within a connected system that knows what to do with that increased speed and flexibility. Otherwise, companies simply become capable of generating high-speed confusion, and running more parallel workstreams that need to be made sense of after the fact.

The companies that will create the most value from AI will not be the ones with the most AI tools. They will be the ones with the clearest vision for how AI supports a fundamentally different way of operating.

The Operating Model Must Come First

Instead of asking yourself whether AI will replace 3D, I believe the more interesting question is what happens when the two technologies are allowed to work together to define a new operating model. That’s a principle we’re proving out with HYPEDROP.

At HYPEDROP, we deliberately avoided treating AI and 3D as separate initiatives. Instead, we built a workflow where each technology contributes at a different stage of the product lifecycle.

AI is primarily used at the beginning of the process to accelerate exploration. Instead of investing days into developing a single concept, teams can rapidly test multiple creative directions, evaluate alternatives, and align around the most promising opportunities.

Once a direction has been selected, the process shifts into 3D. This is where styles move from possibilities into products. Fit, proportions, construction details, and production feasibility can be reviewed, challenged, and approved digitally before physical production begins.

Once products have been approved, the workflow shifts again. The same digital assets that supported development can then be used to accelerate marketing, sales presentations, product storytelling, and digital selling experiences through AI-enabled visualisation.

What emerged from this process was not an AI strategy or a 3D strategy. It was a digital decision-making framework where technology serves the workflow rather than the other way around. The practical outcome is that we can move from initial idea generation to a product that is approved for production and ready to be sold digitally in approximately two weeks. Instead of making decisions based on forecasts for what customers might want six to eight months into the future, we can respond much closer to demand as it exists today.

In an industry that has spent decades trying to predict the future, that ability to shorten the distance between demand signals and product decisions may ultimately be more valuable than any individual technology.

In a digitally-native product development workflow like the one we’re exploring, AI can dramatically accelerate the earliest stages of the process. Ideas can be explored faster. Multiple directions can be visualised within minutes. Design teams can test concepts, challenge assumptions, and evaluate a broader range of opportunities without committing significant time or resources. 

Historically, much of this exploration either happened manually or required physical development work to move ideas forward. AI changes that equation by allowing a much larger volume of possibilities to be explored before physical commitment is required.

However, exploration alone does not create a product.

At some point, ideas need to move from possibility into certainty. Decisions need to be made. Products need to be approved. Suppliers need to understand exactly what is being manufactured. This is where I believe the industry continues to underestimate the role of 3D, and it’s where we’re going to find ourselves looping back around to that original promise of a digital twin, leaving the companies that abandoned it feeling a sense of regret.

Instead of waiting weeks for physical samples to move between suppliers, offices, fit sessions, revisions, and approvals, teams that combine 3D with AI can review, adjust, align, and approve products digitally within hours or days. Development has the potential to become less dependent on physical movement and more dependent on digital information, so decisions (the correct ones) can be made earlier. 

Once products have been validated, those same digital assets can continue creating value throughout the rest of the lifecycle. AI can then be reintroduced to accelerate visualisation, marketing content, sales materials, and customer-facing experiences, all based on products that have already been approved for production rather than speculative concepts.

In this model, AI accelerates exploration. 3D validates decisions. AI then accelerates communication and commercialisation.The result is not an AI workflow or a 3D workflow. It is a connected digital workflow where different technologies support different stages of the process, and where that process can actually be re-engineered rather than just tweaked.

Not because AI makes fashion faster. Not because 3D reduces samples. But because together they create the conditions for an operating model that is inherently more agile, scalable, and adaptable than the physical-first models the industry has traditionally relied upon.

The Future Belongs To Adaptable Businesses

AI will not be the last technology to promise transformation, just as 3D was not the first. Another tool, platform, interface, or intelligence layer will arrive after AI (whether it’s AGI or not, time will tell) and the industry will almost certainly be tempted to repeat the same cycle again.

The winners, in the meantime, will not be the companies that correctly guess which technology becomes dominant. They will be the companies that build business models flexible enough to absorb new technologies without needing to rebuild themselves every time the market changes.

The future of fashion will not be decided by the technologies we adopt. It will be decided by whether we are willing to build businesses capable of taking advantage of them.