More thoughts on technical debt

Suppose you spend years working on an engineering problem. Every day is a sprint, and you end up using the hot glue gun a little more than anyone is comfortable with. Your product barely sticks together and your engineers need to spend time every day propping it up. You have a technical debt problem. Every day, you pay interest on your sloppy or rushed work from before.
OK, now that we know what technical debt is, let's talk about how much it costs.

The business perspective

The fundamental health of a business is revenue minus expenses. The bottom line is that if technical debt isn't directly and obviously contributing to customer-facing bugs, it doesn't stop the product from being sold, and the finances stay in the black.
So you are told that technical debt is starting to build up and become a problem. Since you aren't empowered to change how your engineering organization does their work, and because you can't spare anyone to stop and fix things, simply hiring an additional programmer to meet with the demands of mounting technical debt seems like Good Business.

The engineer perspective

You, the engineer, are anxious about technical debt because it impacts your ability to get things done. If you don't address the debt, your work proceeds more slowly and it has more flaws.
At the same time, if you spend time "paying down" the debt, you might be accused of not spending your time well. Even under the best of circumstances, working on technical debt doesn't feel productive, because you're not doing new things, you're just making old shitty things work better.

The real cost of technical debt

Technical debt brings your engineers down. If you never schedule time for your engineers to address technical debt, their daily stress level increases until you have a big morale problem. People start to just walk away from their (well-paying!) jobs.
Technical debt contributes to the loss of clarity into how your shit works, which will lead to increasingly poor outcomes in your engineering team as they become incapable of understanding how things work, much less fixing them.

Pay down 75% of the debt

If engineers get to fix 100% of what they want to fix, it is my experience that indeed, development will come to a standstill.
So just work on the easiest 75%. If you have seasoned engineers, they will be comfortable with tradeoffs. They can come up with a plan to pay down 75% of the debt for a a relatively small cost, compared to the cost of paying down 100% of the debt.