AI makes a process faster. It also makes an unclear process fail faster. Fix the system first.
AI tools are genuinely useful. I use them every day — for research, drafts, analysis and building small systems. But I've noticed a pattern: teams reach for AI to fix problems that aren't about speed.
Automation amplifies whatever is already there. A clear process becomes faster. An unclear one produces confusion at scale.
Three questions before you automate
- Can someone explain this process in five steps? If not, the problem is design, not effort.
- Do we agree on what "good" looks like? AI can produce endless output; only a clear standard tells you which output to keep.
- What decision does this process support? If nobody uses the result to decide anything, automating it just creates more noise.
Where AI shines
- Turning a clear standard into first drafts the team can refine.
- Summarising large amounts of information around a specific question.
- Connecting tools so people spend time on judgement instead of copy-paste.
Start with the system: the steps, the standard, the decision. Then let technology carry the weight.
A question to leave you with
Which process would you be embarrassed to explain step by step?
Write back to me