You know that system.
The one running your supply chain. Your clinical workflows. Your entire operations backbone.
Built in 2007. Maybe earlier. Works fine. Mostly. Nobody really understands the code anymore because the guy who wrote it retired in 2019. And every time someone suggests modernizing it, the room gets very quiet.
"It's not broken."
"We can't afford the downtime."
"Remember what happened to [competitor] when they tried?"
Here's the thing: 62% of organizations are still running legacy systems. And 70% of enterprise modernization projects fail.
So the fear isn't irrational. It's pattern recognition.
But here's what that fear is actually costing you: $2.41 trillion annually in poor software quality across US companies. $1.52 trillion in accumulated technical debt. And in some sectors, 80% of IT budgets just keeping the lights on.
That's not maintenance. That's a hostage situation.
Why 70% of Modernization Projects Fail
1. They modernize the technology without modernizing the trust.
New system. Same black-box behavior. Same "just trust the algorithm" approach. Users reject it within weeks.
US trust in AI has dropped to 32%, according to the 2025 Edelman Trust Barometer. You can't fix a trust problem with better technology. You fix it with transparency.
2. They try to boil the ocean.
Full replacement. 18-month timeline. Everyone involved. By month 8, the project is over budget, behind schedule, and the champion who started it has moved to a different role.
3. They optimize for IT, not for users.
The system works perfectly in the demo environment. Goes live. Adoption: 11%.
Not because users hate change. Because the interface doesn't fit how they actually work. A 78-minute workflow that could be 8 minutes. But nobody asked the anesthesiologist.
4. They underestimate the "it still works" inertia.
50% of organizations say "current system still works" is the biggest blocker to modernization. And they're right. It does work. Today.
But legacy systems are like that car you've been driving since 2008. It runs. It gets you there. But the parts are getting harder to find, the mechanic who understands it is retiring, and one day it's going to leave you stranded somewhere inconvenient.
How the 30% Get It Right
The organizations that successfully modernize share a few things:
They start with trust architecture, not feature lists.
One healthcare system invested $2M in an AI diagnostic tool. Internal team spent 18 months on the interface. Physician adoption: 11%.
We redesigned it. 68% adoption in 4 months. Same AI. Same hospital. Same physicians who "hate new systems."
The difference? We made the AI show its work. Every recommendation came with the reasoning. Physicians could see why the system suggested what it suggested.
Not intuitive. Defensible.
They modernize in phases, not explosions.
The supply chain leaders winning right now aren't replacing everything at once. They're building capability layer by layer. Real-time visibility first. Then inventory optimization. Then supplier performance management. Then predictive analytics.
Each phase proves value before the next one starts. Each success builds political capital for the next investment.
They design for the skeptic, not the champion.
Your pilot users love the new system. They were hand-picked. They got white-glove support. They wanted it to succeed.
Your average user? Skeptical. Busy. Looking for reasons to stick with what they know.
The 30% design for that person. Not the early adopter. The late majority who makes or breaks adoption.
They have the right teams, time-boxed missions, and guardrails.
Legacy modernization isn't just a technical challenge. It requires:
- Teams with both domain expertise AND design thinking
- Clear, time-boxed missions (not endless "digital transformation" programs)
- Guardrails that keep people focused on delivering real value
- Research that connects to actual user workflows, not boardroom assumptions
The organizations that get this right treat modernization like a series of focused sprints, not a marathon with no finish line.
About the author
Founder, Thirty Two North. 15 years enterprise UX. 200+ complex B2B products across healthcare IT, fintech, logistics, and enterprise SaaS. 6 iF Design Awards. Pattern recognition is the service.
Follow on LinkedInJoin the conversation
This article was originally published on LinkedIn. Share your thoughts, challenge assumptions, or connect with the author there.
Comment on LinkedInMore from this series
Edition 01: They Built Exactly What You Asked For
Why clinicians can't tell you what they need. The gap between stated requirements and actual workflow.
Edition 02: Your AI Is Smarter Than Your Doctors
94% AI accuracy. 80% physician accuracy. 11% adoption. The defensibility problem in healthcare AI UX.
Edition 03: The Training Lie
$400K in training bought 3 percentage points. $120K in interface redesign bought 39. The math that exposes the training excuse.