Technical debt isn't a moral failing—it's a portfolio problem. I'd manage it like risk and capacity, not a guilt trip in retro.
How I'd reduce it without freezing the roadmap
- Make debt visible: inventory with owners, impact, and trigger conditions
- Tie paydown to product outcomes (faster delivery, fewer incidents, safer changes)
- Budget a steady tax (e.g., a slice of each sprint) plus occasional focused campaigns
- Stop adding high-interest debt for low-value features
- Celebrate killed complexity, not only new features
What I'd prioritize first
| Debt type | Why it jumps the queue |
|---|---|
| Blocks customer-facing reliability | Trust and churn |
| Slows every feature behind it | Multiplicative cost |
| Hides in a single hero's head | Bus-factor risk |
I'd avoid "rewrite everything" as the default. Thin wedges beat big-bang fantasies.
A Technical PM Certification helps you partner on these calls. More in the newsletter.