The Challenger space shuttle had a process. An extensive, well-documented, multi-layered process for safety review. Engineers at Morton Thiokol raised concerns about the O-ring seals in cold temperatures. The process received the information. The process escalated it. The process considered it.
The process killed seven people.
Not because the process was poorly designed. Because the process had become a substitute for judgement. The engineers who knew the shuttle should not launch were overridden by managers who knew the schedule could not slip. The process gave the managers cover. They could point to it afterwards and say: we followed the procedure.
Process does not think. That is precisely its appeal.
Consider an aircraft safety inspector working through a checklist. In a growth mindset culture, they don’t just follow the procedure — they stay curious, alert to anomalies, attentive to details not captured by the standardised process. In a protective culture, they tick the box and move on.
Same person. Same checklist. Completely different outcome.
The difference is not the process. It is whether the culture around the process makes curiosity safe or makes compliance sufficient.
The Columbia Accident Investigation Board found the same pattern seventeen years after Challenger. NASA had better processes, more reviews, more data. The foam strike that killed the Columbia crew was identified during the mission. The process considered it. The process decided it was within acceptable parameters. The parameters were wrong. But the process had spoken, and nobody with the authority to override it felt safe enough to do so.
Organisations love process for the same reason leaders love control. It creates the feeling of certainty in an uncertain world. It provides a structure that can be audited, measured, and pointed to when things go wrong.
But the things that actually go wrong — the slow accumulation of unspoken concerns, the gradual normalisation of deviance, the erosion of curiosity under the weight of compliance — those don’t show up in a process audit. They show up in the conversations that nobody had.
The question is never “do we have a great process?” It is: “what is the process preventing us from seeing?”
If nobody has challenged your process recently, that is not evidence that it works. It is evidence that nobody feels safe enough to challenge it.