This is what I personally like, high-level, to advocate when it comes to tech debt:
1. Have solid principles in place that prevent unnecessarily adding to it;
2. Should you have to take a shortcut -- clean it up the next opportunity you have (here's where reserving some capacity / space for that will pay off in the long run);
3. Use tech debt as gap filling of priorities for when that's reasonable (e.g., you are yet gathering more feedback to validate your latest launch, maybe that's an opportunity to let engineering tackle some tech debt they think should be dealt with now)
This is what I personally like, high-level, to advocate when it comes to tech debt:
1. Have solid principles in place that prevent unnecessarily adding to it;
2. Should you have to take a shortcut -- clean it up the next opportunity you have (here's where reserving some capacity / space for that will pay off in the long run);
3. Use tech debt as gap filling of priorities for when that's reasonable (e.g., you are yet gathering more feedback to validate your latest launch, maybe that's an opportunity to let engineering tackle some tech debt they think should be dealt with now)