It's not just legacy tech. It's legacy thinking.
Thinking formed in practice, published as part of the Bearing & Course Points of View library.
One of the persistent myths in technology is that the biggest, best-resourced, fastest-moving organisations have somehow solved tech debt. That once you reach a certain level of scale or financial success, it magically disappears. It doesn't.
Even the most well-resourced organisations carry tech debt. Often a lot of it. When you are growing fast, chasing customers, launching products and trying to stay ahead of the market, trade-offs get made. Documentation gets skipped. Temporary fixes stay longer than planned. Architecture decisions get postponed so something can ship now rather than later. That is not failure. It is momentum catching up with you.
The pattern looks different depending on the organisation. In fast-growth businesses, tech debt is the cost of speed. In more established organisations, across every sector, it tends to accumulate differently. Systems that were well-built for their time are now being asked to do things they were never designed for, in a world that has changed around them. The infrastructure holds, mostly. But the cost of keeping it holding keeps rising.
That cost is not only financial. It shows in where capable people spend their time. Teams that should be building spend their days reacting to incidents, patching issues and working around limitations. It shows in what organisations cannot do: the reforms they cannot implement quickly, the customer experiences they cannot improve, the decisions they cannot make at speed because the data is not clean or the systems cannot talk to each other.
The consequences land on real people. A customer whose claim takes weeks longer than it should. An employee whose onboarding is fragmented and manual. A patient whose records are not available at the point of care. A family whose application stalls because two systems do not integrate. Every workaround, every manual step, every patch quietly chips away at trust: in the organisation, in the service, in the institution behind it.
Legacy technology risks appear on risk registers across every sector. They are flagged year after year, often treated as too complex or too disruptive to address directly. That caution compounds the problem. The longer the delay, the more expensive and fragile the situation becomes.
There is also a gap between what many organisations say they want and where the investment actually goes. The language of modern engineering, product thinking and continuous improvement is common. The reality is that the majority of technology spending still goes to sustaining what already exists. When more than half of your investment is spent maintaining fragility rather than building capability, the aspiration and the practice are not aligned.
You cannot move toward modern without creating space to breathe first.
Modernising legacy systems does not require a years-long overhaul. Done well, it can be tackled incrementally, with real value delivered early and often. The right approach isolates fragile components, builds new capability alongside the old, and lets the new gradually take over. It proves the concept before scaling it. It does not break what still works while building what comes next.
This is not a technical problem in isolation. Legacy lives in governance structures, funding cycles and procurement models just as much as it lives in code. Organisations that address the technology without changing the thinking around it tend to find themselves in the same position within a few years, with a different set of systems that are already accumulating debt.
The tradeoff is not reckless. A modest, well-placed investment now reduces operating costs, improves responsiveness and builds the confidence to keep going. The alternative is continuing to spend heavily on sustaining environments that are becoming more fragile and more expensive every year.
Legacy might describe where many systems are today. But legacy thinking is what determines whether they stay there.