For those of you who started that AI pilot end of Q1: how's adoption going? Is your team still using it at all?
I think of this as the 60-day cliff.
Most AI pilots and projects die in the weeks after kick-off. When nobody's watching anymore. The budget's already been discussed. Attention's moved to other urgent meetings.
Failure patterns are the same whether you're in a team of five or five hundred:
1. No named owner, so the pilot depends on one person's enthusiasm
2. A use case too small for anyone to notice
3. A playbook nobody updates after week one
4. Adoption stalling at "good enough", then fatigue setting in
5. No baseline captured, so nobody can prove what changed
6. Results will never translate into a number the business even cares about
Not one of these is a purely tech problem. That's the part I find genuinely exciting: everything on that list can be fixed with an honest (and maybe even hard) conversation you can have this week.
Which one do you think is hardest to prevent, even with a strong team?