Quiet under load.
Technical debt at Fortune 200 logistics scale. Millions of shipments, real-time coordination, zero tolerance for failure. The work that became Tech-Debt Triage.
Modernise the platform that can never stop running.
CH Robinson operates one of the world's largest logistics platforms. Millions of shipments, real-time coordination, zero tolerance for failure. The challenge: how do you modernise a platform that can never stop running?
The team was caught in a cycle. Incidents consumed engineering bandwidth. Bandwidth gone meant debt couldn't be remediated. Unremediated debt caused more incidents. Breaking the cycle required treating technical debt as a business decision, not an engineering preference.
A matrix, applied four ways.
Triage every piece of debt by business impact and effort. Each quadrant gets one action, and a line item leadership can read.
Quick wins
Items resolvable quickly with significant operational impact. These freed up capacity for larger initiatives.
Strategic projects
Major architectural improvements scheduled for dedicated sprints. Justified to leadership with projected incident reduction.
Opportunistic fixes
Small improvements addressed when engineers were already in affected areas. No dedicated time, continuous incremental improvement.
Accepted debt
Some debt isn't worth fixing. Explicitly documented as accepted, with triggers defined for when re-evaluation would be warranted.
Where Tech-Debt Triage came from.
This engagement directly informed the Tech-Debt Triage framework. The key insight: technical debt prioritisation is a business decision that requires business-level visibility and accountability.
The framework establishes a shared language between engineering and business leadership. Impact is measured in business terms, incident frequency, velocity constraints, opportunity cost, and the result is rational prioritisation rather than engineering preference.
"Technical debt isn't an engineering problem. Left unaddressed, it constrains revenue, slows product, and limits your options at exactly the wrong time. The question isn't whether to address it. It's how to prioritise based on business impact."
Jędrzej Tabaczyński
What stayed.
Numbers are useful. The structures below are what the team carries now.
~40% Sev-1 incidents
Production incidents reduced by forty percent within six months. Strategic debt remediation eliminated the root causes of recurring issues.
~2× developer velocity
Engineering velocity effectively doubled as teams spent less time fighting fires and more time building features.
A shared business vocabulary
Engineering and business leadership built a common language for technical debt impact. Investment decisions became data-driven.
Is the cycle starting to look familiar?
Incidents consume capacity. Debt grows. More incidents follow. The framework is straightforward, the harder work is organisational alignment.