I spend a lot of time with operators running businesses between $10M and $200M. The AI conversation in those rooms sounds nothing like the one happening on Twitter or in keynotes.
There is no debate about AGI timelines. There is a controller closing the books two days faster, a sales lead drafting five proposals in an afternoon, and a founder finally answering every customer email the same day. Boring, durable, compounding.
The real unlock is not the model
The teams that are pulling ahead are not the ones with access to the best model. They have the same tools everyone else does. What is different is that a senior leader insists the tools actually get used, every day, for the work that matters.
Adoption is a leadership problem dressed up as a technology problem. Solve it that way.
Where it shows up first
Finance and operations adopt before sales and marketing, almost without exception. The work is more structured, the prompts are reusable, and the upside is immediate. A monthly close that drops from twelve days to seven is not a hypothetical — it is the median outcome I see in the first six months.
Customer support comes next. Sales takes the longest, partly because the work is more relational and partly because compensation systems take time to update.
Failure modes I see weekly
Buying tools nobody is trained on. Naming a chief AI officer with no operational authority. Running pilots in isolated teams that never propagate. Letting individual contributors decide which tools to use, with no shared workflow.
Every one of these failures has the same root: nobody senior is treating adoption as a measurable program with weekly check-ins.
What to do on Monday
Pick one workflow per function. Assign one person to own it. Measure cycle time before and after. Review the numbers monthly with the same seriousness you bring to revenue.
Do that for two quarters and the cumulative effect is the equivalent of hiring ten people who never get tired and never ask for a raise. It is the most leveraged operating move available to a mid-market leader today.
The competitive gap is not opening between companies that have AI and companies that don't. It is opening between companies that use it and companies that pretend they do.