Technical Debt is a Risk, not an Evil

It's surprising how often technical debt is associated with pure evil; as something to avoid at all costs. I believe, however, that technical debt is entirely analogous to financial debt and since most people have a solid intuitive understanding of money, I find it helpful to discuss technical debt in this context.

Crippling Debt

Uncontrolled or impulsive spending on your credit card will invariably land you in trouble. Failing to consider the implications of interest obligations on a growing, outstanding credit balance leads to pain and, in the worst cases, lifelong indebtedness (it's really enslavement) or bankruptcy.

Similarly, suppose you keep making bad technical decisions without considering the impact on your solution's architecture. As a result, you're racking up maintainability, performance and extensibility debt with no regard to the future, soon finding yourself in a position where you can no longer service the interest, let alone pay back the capital. Your architecture calcifies, and the fear of disastrous regression impacts accompanies even the most conceptually minor feature requests or changes to existing functionality.

This is the category I believe most people automatically assign to technical debt and rightfully fear, but it does not have to be this way.

Debt as Leverage

Another type of debt is a calculated, considered loan that serves your interests. Of course, there is still a risk that the debt can lead to an unfavourable outcome, but you consciously engage with the problem and judge the risk acceptable for the potential gains you foresee.

For example, very few of us have the means to buy a property outright. However, assuming you do your due diligence and consider your ability to service the interest, weather rate hikes, etc., a mortgage could ultimately lead to owning your own property while providing accommodation in the process. Another example is a revolving credit facility you can lean on when cash flow is undesirable. Of course, the preference would be not to dip into the credit. Still, a potentially far more significant loss would be your employees walking out the door when their salaries don't arrive on time (hopefully it's also apparent that paying people on time is the right thing to do).

Analogously, in the software business, there are scenarios where taking a hit w.r.t. technical debt could be beneficial. In my experience, most of these scenarios relate to time to market.

Shortcuts, duct tape, “swamp donkey hacks”, call it what you will. You don't need to like it, and should always strive to be debt free, but the option to take on considered debt to get to an ROI faster should always be on the table.