“Make everyone one percent better and it compounds across the team.”

That’s the line Joe Head wrote about his own job and when I read it back to him, he didn’t hesitate. “That is 100% the objective,” he said.

Joe runs AI adoption for an 80-person team as the only AI automation specialist in the business. Part of the job is building automations, part of it is training people, and he protects that one-line philosophy from something most AI strategies never account for… the tool itself.

Build the skill, not the tool habit

Ask most MSPs how they’re rolling AI out right now and you’ll hear some version of the same answer. Get the team good at Claude. Get the team good at Copilot.

Joe thinks that’s the wrong bet and not because the tools are weak.

“We should never focus on the tools, because the tools change all the time,” he told me. “In a year’s time, that will change because the model changes and there’s something new.”

His answer is to build the strategy around the person instead. What matters is the person’s ability to adapt, because that’s what doesn’t reset every time a new model ships. It’s not “can you prompt Claude.” It’s whether someone can tell what’s worth automating in the first place.

If we can give our teams the skills for the way the world is going to work in a few years’ time, that’s a much stronger vision than saying we’re going to make sure everyone can use ChatGPT, which will change anyway.

If you name the tool in your AI strategy today, you’ve written an expiry date into it.

AI didn’t come with a playbook

Joe’s confidence here doesn’t come from having it all worked out. It comes from admitting nobody does.

We don’t have enough history of delivering this kind of transformation to look back at the mistakes that businesses have made,” he said. “Everyone’s figuring it out. No one knows the best way to do it.

Compare that with a financial transformation project. There’s a hundred years of precedent to draw on with a track record of what’s worked and what’s failed, while AI transformation has none of that.

Every business rolling this out is guessing right now, Joe included. He makes peace with that by building the strategy around people instead of tools.

He puts the underlying point more simply elsewhere in our conversation. AI isn’t a phase. “It’s the same as email,” he said. “It will be around pretty much forever.” That’s exactly why treating it as a skill worth building beats treating it as a tool worth mastering. One of those outlasts the other.

That gives any MSP leader a genuinely useful test to run before booking the next training session. Are you teaching a tool, or building a skill your team keeps regardless of which tool wins next year?

For Joe, that shows up in a simple detail. Training isn’t a separate initiative bolted onto the work. It’s a byproduct of the automation itself. He builds it in as he goes rather than delivering it as a separate course.

Which brings this back round to where it started. One percent better. That compounds across the team, whatever happens to be powering the work next year.

Joe also had a clear rule for what happens when speed replaces judgement in how these tools get used. That’s a conversation worth its own installment.

If you liked this article, follow us on LinkedIn, Reddit, X, Facebook, and Youtube.