Architecture
What’s Below the Waterline
Apr 25, 2025
“Tech debt is a measure of the amount of duct tape holding your system together, plus the amount of rust that it has accumulated.” —Itzy Sabo
Tech debt isn’t just a technical problem anymore—it’s a leadership problem. In boardrooms and strategy meetings, it needs to be as much of a priority as revenue or market share.
AI is reshaping industries faster than organizations can keep up with. Companies that once tolerated slow, brittle systems now face a harsh reality: if your foundation can’t support AI, your future is already compromised.
Cutting corners today always haunts you tomorrow. After years of leading engineering teams, one lesson stands out: technical integrity isn’t a luxury—it’s the deciding factor in how fast and far you can go.
Why Tech Debt Matters More Than Ever
Tech debt accounts for 20–40% of the typical enterprise IT budget, slowing innovation and compounding long-term costs. I’ve seen it firsthand: in one case, nearly a third of our development time was wasted on maintenance workarounds instead of building new customer features.
Think of tech debt as a high-interest loan. Every shortcut, every “we’ll fix it later,” and every legacy system kept on life support accrues interest in the form of:
Slower development velocity
Rising maintenance costs
Frustrated engineers
Security vulnerabilities
Competitive disadvantage
And the bill always comes due—usually when you can least afford it.
When Technical Shortcuts Become Strategic Barriers
In the AI era, tech debt doesn’t just slow you down—it blocks your ability to compete.
Organizations with well-maintained, modular systems can launch AI initiatives within months. Teams drowning in tech debt spend years and still deliver far less. According to McKinsey, companies should address tech debt before pursuing AI projects, or AI’s complexity will only compound their existing problems.
Organizations investing in technical health are better positioned to leverage AI innovations. Those with cleaner architectures can implement AI solutions more quickly, achieve higher-quality outcomes, and incur lower overall costs.
If your systems are brittle, AI won’t save you—it’ll expose you.
Beyond the Big Rewrite
Earlier in my career, I thought the answer to tech debt was a complete rewrite. It sounded good on paper. In reality, that project took twice as long and delivered barely half the expected value.
A better approach is to manage tech debt like a portfolio, rather than treating it as a one-time fix. ThoughtWorks stresses that managing tech debt effectively needs commitment from the whole organization and smart prioritization. It’s not just about fixing code from the engineering team.
According to Martin Fowler’s Tech Debt Quadrant, debt generally falls into four categories:
Strategic: intentional and prudent, taken on to gain early speed.
Tactical: deliberate but potentially reckless, requiring scheduled cleanup.
Inadvertent: unintentional, caused by mistakes discovered after release.
Environmental: results from shifts in technology or business context, often outside direct control.
Each type requires a different strategy. Treating them all the same is a waste of time and money.
Strategic Speed vs. Tech Debt
There’s an important balance to strike between velocity and quality. The most effective organizations distinguish between strategic debt (deliberately taken on with a clear remediation plan) and unplanned debt (accumulated unintentionally or without oversight).
Successful leaders recognize that technical integrity and speed aren’t opposing forces—they’re complementary. Well-architected systems enable faster innovation in the long run, while poorly maintained ones create an increasing drag on new initiatives.
When teams have clarity about which technical compromises they are making and why, they can make informed decisions that serve both immediate business needs and long-term sustainability.
A Smarter Way to Prioritize
So, how do you decide which debt to tackle first? Good tech debt management starts with deliberate focus. I use a simple decision matrix from Accenture called PAID:
Prioritize: High business impact, reasonable effort—fix immediately.
Address: High impact, high effort—make a plan and do it.
Investigate: Low impact, reasonable effort—quick wins worth pursuing.
Document: Low impact, high effort—defer and monitor (address later if it aligns with other work).
For example, one of my teams used PAID to modernize our customer data platform—a high-impact project—instead of a low-value internal rewrite. That decision unlocked far more value for the business. This framework also forces teams to make data-driven decisions about tech debt, rather than relying on gut feelings or deferring to the loudest voice in the room.
Effort alone doesn’t deliver results. Prioritization does.
Tech Debt as a Value Driver
Despite its lack of immediate appeal, cleaning our API layer resulted in a substantial 40% improvement in the time to market for new features. Cloud platforms like AWS’s Migration Evaluator and Google’s Migration Center offer helpful tools to measure tech debt. However, actual change stems from leadership committed to understanding and acting on that information.
Building a Culture That Prioritizes Technical Health
No tool or technology will fix tech debt. Only culture can.
Harvard Business Review puts it bluntly: tech debt, like process debt, requires leadership intervention. In other words, leaders must actively engage with and understand tech debt if they expect to prioritize it properly.
In the best teams I’ve seen, leaders:
Make tech debt visible through dashboards and status reports.
Dedicate about 15–20% of engineering capacity to addressing it.
Reward teams for paying it down, not just for shipping new features.
Bake tech debt review into the definition of “done.”
Continuously educate business stakeholders on why it matters.
Healthy systems don’t happen by accident; they result from steady investment.
Foundations Matter More Than Ambitions
As AI reshapes the playing field, your technical foundation—not your aspirations—will determine your future. McKinsey’s research on the business value of technical readiness confirms that companies with strong technical practices outperform their competitors in almost every important business metric.
Organizations prioritizing technical agility will best seize new opportunities as AI changes the competitive landscape. Those who systematically manage tech debt will maintain the agility needed to adapt and innovate.
Consider measuring your “Innovation Ratio”: What percentage of your engineering capacity is dedicated to building new capabilities versus maintaining existing systems? This single metric can provide valuable insight into your organization’s technical health and readiness for AI transformation.
In today’s AI-driven world, managing tech debt isn’t optional—it’s existential. Teams that ignore it won’t just move slower—they’ll miss the next wave entirely.
