Skip to main content

Tech-debt audit method

A repeatable, per-product method to find the debt that actually hurts — not every imperfection, but the debt with real interest. Runs during STOP; feeds the tech-debt register and the remediation backlog.

The method​

  1. Static analysis — quality, complexity, duplication (ISO/IEC 5055 where applicable).
  2. Churn-vs-complexity hotspots — cross-reference change frequency with complexity to find debt that's both fragile and frequently touched. This is where debt costs real money.
  3. Dependency health — outdated, unmaintained, or vulnerable dependencies (feeds the EOL inventory).
  4. Build/deploy friction — manual steps, slow pipelines, flaky deploys (overlaps the DORA baseline).
  5. Doc gaps — undocumented systems the code-archaeology skill surfaces.

Sizing debt: principal vs interest​

Every finding is sized two ways so items compare across products:

  • Principal — the one-off cost to fix it.
  • Interest — the ongoing drag it causes (slower changes, more defects, more incidents).

High-interest debt is remediated first regardless of principal — it's the debt paying negative dividends every sprint.

Output​

Every finding lands in the tech-debt register with evidence, principal, interest and a proposed remediation — or it doesn't exist.

Standards referenced: ISO/IEC 5055, SQALE, C4.