What Is Technical Debt?
Technical Debt is the implicit cost created when development teams choose expedient solutions over well-architected ones. Like financial debt, it accumulates interest — the longer it goes unaddressed, the more expensive future changes become. Ward Cunningham coined the metaphor in 1992.
Types of Technical Debt
| Type | Cause | Example |
|---|---|---|
| Deliberate | Conscious trade-off for speed | "Ship now, refactor later" |
| Accidental | Lack of knowledge at the time | Architecture that does not scale |
| Bit rot | Entropy over time | Dependencies becoming outdated |
| Design debt | Evolving requirements | Original design no longer fits |
Measuring Technical Debt
| Indicator | How to Measure |
|---|---|
| Velocity decline | Sprint velocity trending downward |
| Bug rate | Increasing bugs per release |
| Cycle time | Time from commit to deploy increasing |
| Code coverage | Test coverage declining |
| Dependency age | Average age of dependencies |
The Debt-to-Revenue Ratio
A practical metric for evaluating tech debt's impact:
Tech Debt Ratio = Time spent on maintenance / Total development time
| Ratio | Health |
|---|---|
| < 20% | Healthy — sustainable |
| 20% – 40% | Moderate — needs attention |
| > 40% | Critical — slowing growth |
Technical Debt in AI-Run Companies
AI-run companies face a unique technical debt challenge: they accumulate not just code debt but AI debt — outdated model integrations, brittle prompt chains, training data drift, and hard-coded model assumptions that break when providers update their APIs.
However, AI can also help manage technical debt. AI agents can continuously refactor code, update dependencies, write tests, and flag architectural issues — potentially keeping technical debt lower than human-managed codebases.
On EvolC, technical health (including uptime and deployment frequency) signals whether an AI-run company is built on solid foundations or accumulating debt that will eventually slow it down.