Prevention is the strategy for performance in infrastructure delivery
From March issue, NZ Manufacturer magazine Stephanie Pretorius, Managing Principal, Argon & Co Infrastructure projects operate under relentless pressure: tightening margins, rising complexity, public scrutiny, and near-zero tolerance for delay. Yet many delivery environments still default to recovery — catch-up programmes, overtime, task-force escalation — only after performance has already deteriorated. In manufacturing, results improve when the production system is stabilised. Once flow is predictable, investment in capability, automation, and quality at source delivers compounding return. Stability creates ROI. Construction rarely affords itself that discipline. Each project is framed as unique, variability is normalised, and management attention focuses on reacting to disruption rather than designing it out. But the underlying principle does not change. Stability precedes return. In infrastructure delivery, prevention — of defects, rework, design ambiguity, coordination failure, and flow breakdown — is the equivalent of investing in a stable production line. It is a deliberate decision to engineer reliability into planning, commercial strategy, stakeholder alignment, and daily management before work intensifies. The most reliable infrastructure projects take a different path. They do not wait for delay to surface. They design systems that prevent instability from taking hold. Prevention is not a support activity. It is the strategy. Why recovery fails as a performance model Most delays are not caused by a single failure or a lack of effort. They are the predictable outcome of how work is planned, coordinated, and managed across the system. Common causes include: Unreliable handoffs between teams and contractors Late decisions and unresolved constraints Recurring problems across projects not addressed as delivery risk Overloaded schedules with no capacity for variation Poor visibility of emerging performance risks Escalation that happens too late—or not at all By the time these issues show up on a critical path report, the cost has already been incurred. Recovery […]
