There's a quiet piece of magic in every business. Your core software — the system that handles orders, invoices, customers, inventory, payroll — does all the heavy lifting so reliably that nobody has to think about it.
Here's the catch: your growth strategy sits on top of that software. Which means your business has to somehow grow using tools designed for the company it used to be.Gartner, 2025 survey of 3,186 technology executives. 62% of strategy leaders say their legacy operating model cannot support current and future strategic objectives.
Gartner's 2025 survey of 3,186 technology executives found that 62% of strategy leaders say their legacy operating model cannot support current and future strategic objectives. The question isn't whether your software is old. It's whether it has crossed into actively constraining growth.
The Route Map Problem §
Here's a scenario to make the distinction concrete. Ludicrous but morally equivalent to what happens in most enterprises.
Imagine running a national delivery operation with maps drawn twenty years ago. Some deliveries get lost on roads the maps don't show. Some addresses are simply wrong, impossible to fulfill.
Now imagine a new service promising real-time tracking: packages arrive on time, to the correct address, every time. Except this tracking service has no way of delivering other than the outdated maps underneath.
So it compensates. Someone manually corrects high-volume routes. Someone cross-references every new address against current data. Someone reconciles the discrepancies by hand each morning. The end customer never hears about the problem.
What "Working Fine" Actually Means §
That is, approximately, the situation with legacy business software, what executives like to call a mature system. A simplification of something much messier going on under the covers.
A lot of running a business consists of building on these simplifications. What is an ERP? A way to pretend that hundreds of interconnected processes can be managed as one coherent system. What is a CRM? A way to pretend that your entire customer history is organized, accessible, and complete.
When your business outgrows the assumptions your software was built around, when markets and data requirements and AI ambitions have moved past what the system was designed for, the software can't close the gap. And the gap widens every quarter.
This is the liability crossover: your core systems attempt to provide a complete abstraction of your operations, but reality leaks through, and you feel the things the abstraction can't quite protect you from. The question shifts from "does it work?" to "what else are we spending to compensate for its limitations?"
All non-trivial business software eventually becomes a constraint on the business it was built to serve. Systems age. There's drift. Things go wrong. It happens across every enterprise running legacy systems, and the signals are consistent.
Where The Leaks Show Up §
Something as simple as deploying an AI initiative can produce radically different outcomes depending on what your core systems actually deliver. The IBM Institute for Business Value reports that 86% of executives say technical debt directly constrains their AI success — and companies that fail to account for legacy costs upfront project 29% lower ROI.IBM Institute for Business Value. 86% of executives say technical debt directly constrains AI success; companies that fail to account for legacy costs upfront project 29% lower ROI.
You're not supposed to care about what's underneath, only what the system delivers. But when the abstraction crumples under the pressure of business reality, workarounds multiply.
Gartner predicts that by 2027, 75% of employees will acquire, modify, or create technology outside IT's visibility, up from 41% in 2022.Gartner, 2027 prediction. 75% of employees will acquire, modify, or create technology outside IT's visibility, up from 41% in 2022. Each workaround is rational on its own. Collectively, they represent a parallel technology stack: invisible, ungoverned, and expensive.
The diagnostic question isn't whether the software works. It's what else you're spending to compensate for its limitations.
And you can't hire your way out. Even with documentation, training programs, and institutional knowledge, the talent pool for decades-old platforms keeps shrinking. ISACA's 2025 survey found that one in three technology professionals switched jobs in the past two years. When your system depends on a dwindling number of people who understand it, that's a concentration risk, regardless of how well it technically runs.
Why Replacing It Isn't Simple Either §
One reason the liability crossover is so difficult: replacement doesn't actually simplify things the way it's supposed to.
In an ideal world the recommendation would be to go straight to modernization. But the VA spent $12.71 billion replacing a legacy health records system, deployed it to only 6 of 170 sites, and a GAO survey found 75% of users said the new system made them less efficient.The VA spent $12.71 billion on a legacy health records replacement, deployed to only 6 of 170 sites. A GAO survey found 75% of users said the new system made them less efficient. Anyone considering a major overhaul needs to reckon with modernization risk first.
The liability crossover means that whenever somebody proposes a big transformation that will make the business ever-so-efficient, you need evidence it actually works — not a vendor's assurance that it will.
Old systems cost you growth. Replacing them doesn't save you from risk. And paradoxically, even as the tools get better, making the right call on legacy software is getting harder.
Before your next strategy session, ask one question: if an investor conducted technical due diligence on your business tomorrow, would they flag your software as an asset or a discount on the purchase price? Private equity firms run exactly this assessment before every acquisition. They look for integration limitations, knowledge concentration, workaround spending, and data accessibility.