You need to diagnose before you fix....
Skipping that step is what makes change initiatives less effictive and more risky, especially because we are talking about behavior change.
I know it's easy to jump into solutions.. its so much easier, and most of those solutions are really good.. but the issue is most are not aligned to the right barriers.
If a new tool is not getting adopte, the instinct may be more training... but training may not be the real barrier, which means you spend money and resources on a good solution that solves no problem.
In my work, a diagnosis is a must... and a good diagnosis is the difference between a good outcome and wasted resources..
So, before you jump to the solution... ask yourself three questions:
Does the system support the change? Do the people know how to do it? Do they want to do it?
Let me elaborate with some examples
If you were hoping for AI Adoption:
-System support? → Is there a clear workflow for when to use AI vs. when not to, or are people expected to just figure it out?
-Know how to do it? → Do employees know how to evaluate whether AI output is good enough to act on in their specific role?
-Want to do it? → Do people see AI as making their work better, or as a step toward replacing them?
In Safety Behavior:
-System support? → Does the schedule give people enough time to follow the procedure correctly, or does the pace quietly punish doing it right?
-Know how to do it? → Do workers know which safety steps they can adapt under time pressure and which ones are non-negotiable?
-Want to do it? → Do people believe following safety protocols actually protects them, or do they see it as just covering the company?
If you're leading a change effort, how often are you doing a deep diagnosis?