Most infrastructure failures are not caused by a lack of effort. They are caused by systems that quietly reward the wrong behavior.
According to Gord Reynolds, this is one of the least comfortable truths in infrastructure delivery. Smart, capable people often do exactly what the system incentivizes them to do, even when those incentives undermine outcomes.
“Most organizations are risk-obsessed and outcome-ignorant,” Reynolds says. “And then they’re surprised by the result.”
Projects do not drift by accident. They drift because delay is often safer than decision, ambiguity is often safer than clarity, and escalation is frequently more dangerous than silence.
Why rational people make irrational project decisions
On large infrastructure programs, incentives shape behavior long before problems surface in the field. People learn quickly what is rewarded and what is punished.
Raising unresolved issues early can be risky. It creates friction. It challenges assumptions. It exposes gaps that leadership may not be ready to confront. Waiting, on the other hand, rarely carries consequences. Problems can be logged, monitored, discussed, and deferred without forcing a decision.
Over time, this dynamic produces predictable behavior. Risks are documented instead of resolved. Issues are escalated sideways instead of upward. Activity is measured, but outcomes are not.
No one intends to sabotage delivery. They are simply responding rationally to the incentives in front of them.
The quiet reward of delay
One of the most damaging incentive failures Reynolds sees is the quiet reward of delay. In many organizations, postponing hard decisions feels prudent. It avoids conflict. It preserves optionality. It keeps everyone technically compliant.
But delay is not neutral.
When decisions are deferred, uncertainty compounds. Dependencies harden. Options narrow. Costs increase. What looked like caution early becomes constrained later.
“The 80/20 rule isn’t about efficiency,” Reynolds says. “It’s about courage. Twenty percent of decisions drive eighty percent of outcomes. Smart leaders focus there. Worried leaders hide in the rest.”
Broken incentive systems encourage leaders to hide in process. They reward thoroughness over effectiveness and consensus over accountability. The result is a project that appears busy and controlled while quietly losing its ability to influence outcomes.
When incentives overpower intent
Reynolds is clear that good intentions are not enough. Systems that reward avoidance will produce avoidance, no matter how capable the people within them.
This is why so many post-mortems sound familiar. The risks were known. The issues were flagged. The warnings were recorded. And yet nothing changed until the consequences became unavoidable.
At that point, urgency replaces agency. Teams react instead of lead. And accountability becomes retrospective rather than operational.
“Don’t admire the problem,” Reynolds says. “Solve it.”
But solving problems requires incentives that support action, not just awareness.
Leadership is designing what behavior survives
For Reynolds, leadership in infrastructure delivery is not about working harder or asking people to care more. It is about designing systems that make the right behavior the safest behavior.
That means asking uncomfortable questions. What gets rewarded here. What gets punished. What happens to the person who forces a hard decision early. And what happens to the person who quietly lets the problem grow.
Incentives are not abstract. They are choices. Leaders decide whether ownership is valued or avoided, whether clarity is rewarded or penalized, and whether outcomes matter more than appearances.
Hard work is common. Discipline and clarity are not.
Courage over comfort
Changing incentives is difficult because it exposes trade-offs leaders would rather avoid. It requires prioritizing outcomes over optics and effectiveness over consensus. It requires focusing on the few decisions that actually matter, even when doing so creates discomfort.
Focus is the real flex. Distraction is the death of scale.
Projects do not succeed because everyone tried their best. They succeed because someone was willing to take responsibility when it mattered, and because the system supported that choice instead of punishing it.
Broken incentives do not just slow projects down. They teach people to stop acting.
Until leaders are willing to redesign what behavior is rewarded, infrastructure delivery will continue to struggle, not because people are incapable, but because the system quietly teaches them not to lead.
