Why 84% of Companies Fail at Digital Transformation. And Why That Number Hasn’t Moved.
By: Frank Merino
Nearly a decade ago, Forbes published a survey finding that 84% of companies fail at digital transformation. Since then we’ve gotten better cloud platforms, better integration tools, and now AI on top of everything. The failure rate hasn’t budged.
That should tell you something. The problem was never the technology.
I’ve spent my career on both sides of this statistic. I’ve watched transformations consume a decade and millions of dollars without ever getting off the ground, and I’ve helped guide transformations that delivered exactly what they promised. The difference between the two has almost nothing to do with which platform you pick. Here’s what I’ve learned.
A Decade of Restarts
Let me tell you about a failure I watched up close.
A company set out to build a digital experience for its end customers. The ambition was genuinely great. The idea was big, bold, and everyone involved fell in love with it. That was the problem.
The project ran for nearly ten years, start to finish, with multiple restarts along the way. It was never phased in a way that could generate momentum. There were no milestone wins, no early deliverables in users’ hands, nothing to point to and say “this part works, let’s build on it.” The big idea was so big that all the little things required to actually get there never happened.
Instead, the company kept investing in new platforms, almost as a way to paper over the failures. When the current approach stalled, the answer was a different tool, a different vendor, another restart.
And here’s the part nobody plans for: over a ten-year project, you will have two or maybe three leadership changes in the areas that matter. When leadership changed, the new leaders inherited a big idea with no milestones and no phase wins to anchor to. So they started over. Every time.
The cost was enormous. This was during our shift to cloud compute, and I watched a $2 million annual budget get consumed by February because the team was building for something too big with improper planning. That led to rewrites just to rein in costs, which led to more delays, which led to more restarts.
A decade. Multiple leadership teams. Millions of dollars. The transformation never kicked off the ground, but it started multiple times over.
What the Successful One Did Differently
Now contrast that with a transformation we were directly involved in that worked.
The success wasn’t built around a product or a platform. It was built around a business objective. The client wanted to change the way they were servicing their customers. They wanted to change their operating model. That desire is what allowed the digital transformation to actually take place, because we weren’t fighting historical models and trying to wrap a “digital transformation” label around the old way of doing things.
Because the objective was a business outcome, we had a way of measuring success. We knew what “done” looked like at each stage. That let us do several things the failed project never could:
- We established a program for ongoing transformation, not a one-time implementation.
- We built meaningful milestones that stacked on each other.
- End users were working with the new capabilities very early on, which meant we were getting real-time feedback instead of building in a vacuum.
- Leadership stayed bought in, because they could see progress against goals rather than chasing a big idea on faith.
When you stop treating transformation like the implementation of a product and start treating it as a change in how you do business, the success factors start falling into place. And once they do, you gain the ability to deliver in smaller milestone chunks.
A Word About “MVP”
A lot of people want to call those milestone chunks an MVP, a minimum viable product. I stay away from that term, and I’d encourage you to as well.
The problem is that everyone’s definition of “minimum” and “viable” is different. What one stakeholder considers minimally viable, another considers minimal and not viable. The term itself creates misalignment. Define your milestones in terms of the business outcome they deliver, not in startup vocabulary that means something different to every person in the room.
The Implementation Trap
I work with mid-market manufacturers, and they get sold “digital transformation” constantly. The most problematic pitches I see are the ones pushing a tech platform, usually a warehouse management system or an MES, as transformation itself.
These are great solutions. But they don’t become digital transformation by implementation. An implementation project is exactly that: an implementation project. If you’re positioning a software product to solve a problem but only addressing it as a project, and not as a way of looking at your business differently, you haven’t transformed anything. You’ve installed something.
Here’s the tell I give every client: if you’re being sold something that’s “live tomorrow,” that’s not transformation. That’s an implementation. What makes transformation transformation is adoption. It’s how your users take it up, how it changes your business. A technology can be capable of changing your business and still sit unadopted inside it. Those are two very different states, and vendors rarely draw the distinction.
Layering Intelligence on Dirty Data
The other pitch I push back on constantly is the analytics layer. BI solutions, and now AI solutions, sold as something you drop on top of your existing applications.
Conceptually it sounds fine. In practice, it ignores three things:
- Your data may be dirty, and probably is.
- Your legacy systems may not be able to talk intelligently to other systems, no matter what you wrap around them.
- There are almost always additional requirements that don’t make it into the sales conversation.
You need a modern operational layer where data is accessible across your solutions, or at least can be made accessible. It doesn’t have to happen at full speed on day one. But you have to have a goal in mind for what you want to see different, and that goal should never be “implement this tool.” Implementing the tool might be part of the overall goal, but it’s a stepping stone, not the destination.
Your First 90 Days, Before You Spend a Dollar
If you’re a manufacturing leader about to kick off a transformation effort, here’s what I’d tell you to do before any platform investment.
Map Your Data
Digital transformation starts with foundational elements, and it’s typically data. Whatever information is important to your organization needs to be accessible. If that data only lives inside an application and can only be reached through that application’s interface, with no API layer or data warehouse behind it, that’s an issue. You can overcome some of those limitations, but you need to understand them early, because they will be your weakest link.
The way to do this is data mapping. Understand what systems you’re using. Understand what data moves from one application to another, and what processes move it. That’s it. There’s no platform investment required at this stage. It’s a pure understanding of what your organization is currently doing.
It’s tempting to say “we’re ripping it all out anyway, so it doesn’t matter what we do today.” That might be true on paper, but it’s not how people work. You have to understand how the business runs today, because transformation will be disruptive to your business, and should be, by the way. But it can’t come at the peril of your current business. You have to be able to keep running while you build, and then make a deliberate decision about when you switch to the new way of doing things. Day-one rip-and-replace is not how business works.
Define Success Criteria That Aren’t Always Dollars
Assuming everything is possible, what would success look like? In manufacturing, that might be equipment efficiency. Units per minute or per hour. A decrease in changeover time when equipment has to be recalibrated for a different run. In food packaging, it might be allergen-specific measures.
Notice that none of those are dollar figures. That’s deliberate. Not everyone in your organization understands a dollar the same way, and different teams see their involvement in a project differently. Success criteria that map to how people actually work are criteria people can actually rally around.
Know Who’s Affected Before They Find Out the Hard Way
How is HR involved? Accounting? Operations? Legal? All of these should be considered when scoping a transformation. It may be that nothing changes for a given team. It may be that something significant does. Either way, they should know beforehand, not when the change lands on them and they’re forced to conform without the preparation everyone else got.
Why the Failure Rate Won’t Budge
So why does the 84% persist after a decade of better tools?
Because tools and technology aren’t what make digital transformation. They aid it. There are two words in “digital transformation,” and everyone obsesses over the first one. It’s the second one that kills projects.
Transformation takes time. I like the health analogy: if your goal is to lose weight, you could cut off a leg. You’d lose a considerable amount of weight and change your body composition. Technically, that’s a transformation. It’s also not helping you. Actually losing weight takes time, care, and effort. You have to change the way you do things. Transformation is exactly the same. You can put a new tool in your tool belt, but that doesn’t mean you’ve digitally transformed.
The other analogy I use is the school bus. If you’re alone in the driver’s seat, you can turn left, turn right, do 90 miles an hour. You’re the only one affected. But if you’re driving a school bus full of people and you make a hard left with no warning, you’re tossing people around. You have to tell them: we’re turning left, brace yourself. We’re turning right, brace yourself. Give people time to brace, or you’re just throwing your organization around the cabin, and that is not conducive to a successful business.
That’s what change management actually is. It’s not “I told you once.” It’s not a 10-point plan announced at a kickoff meeting. It’s giving the people who have to do the work the time and attention necessary to adopt the new way of doing things. Not just training. Using it. Adopting it.
The Hard Truth
Digital transformation is hard, and you have to be ready to make decisions that will disrupt your own business. That disruption isn’t a side effect to be minimized. It’s kind of the objective. If nothing about how you operate changes, you didn’t transform anything.
But disruption without change management is catastrophe. The companies that succeed have leadership that understands both halves of that sentence: willing to disrupt, disciplined enough to prepare people for it, and patient enough to know that transformation is not a one-year project. It can take several years, and it requires a lot of care and feeding.
I suspect the failure rate will stay high, because that kind of leadership is rarer than any technology. The good news is that it’s also entirely learnable. Start with a business objective instead of a big idea. Map your data before you spend a dollar. Deliver in milestones people can see. And warn the bus before you turn.
Digital Transformation Starts With Strategy
Technology is only one piece of the equation.
DigitalNOW helps organizations align people, processes, data, and technology around measurable business outcomes, so transformation becomes more than another software implementation.
Whether you’re exploring AI, system integration, automation, or operational modernization, we’ll help you build a roadmap that delivers lasting results.
Ready to transform your business, not just your technology?
→ Check out this case study on Closet Factory.
