I had a moment last week that reminded me just how much work we still have to do in getting our teams up to speed on AI.
I sat down with a VP Finance I’ve worked with for years. Smart guy. Strong background, knows the classic tools cold. Been in SaaS for a decade. Manages a team of 8.
I showed him how Claude works for financial modeling.
His mind was utterly blown. He had no idea.
Not in a “oh that’s neat” way. In a genuine “wait, I can do THAT?” way.
Here’s what hit me: We’re still incredibly early to AI adoption, especially in core business functions.
The Reality Check
This wasn’t some technophobe. This was a VP Finance at a high-growth B2B company who:
- Uses Excel daily
- Builds complex models
- Manages forecasts and board decks
- Considers himself tech-savvy
And yet he had never actually used Claude (or ChatGPT, or any LLM) for his actual work. He’d heard about it. Read articles. Seen the demos. But never sat down and actually incorporated it into his workflow.
When I showed him how to:
- Iterate on model assumptions in real-time
- Generate different scenario analyses in seconds
- Debug formula errors by just pasting them in
- Build out sensitivity tables with natural language
…it changed how he thought about his job.
The Follow-Up (This Is The Part That Matters)
Now? Every single week he sends me his financial modeling work from Claude.
Every. Week.
He loves it now. Can’t stop talking about it. Texts me screenshots of outputs. Shares examples with his team.
This is the same person who, three weeks ago, had never opened Claude.
The transformation wasn’t about the technology. The technology was already there. It was about someone taking 20 minutes to show him how it applied to his actual work.
The Opportunity (And The Problem)
Here’s the thing: Your team is probably in the same boat.
Even at companies that consider themselves “AI-forward,” most employees are still:
- Not using AI tools daily
- Not sure how to apply them to their specific roles
- Intimidated or unclear on where to start
- Defaulting to old workflows because they “work fine”
And that’s a massive missed opportunity.
Because the productivity gains aren’t theoretical. They’re real. They’re happening now. But only for the people who are actually using these tools.
What You Should Do About It
Don’t assume your team is figuring this out on their own. They’re not.
Three things that actually work:
1. Show, don’t tell. Sit down with your team members individually. Show them how AI applies to their specific workflows. Not generic use cases. Their actual daily work. Spend 20-30 minutes. That’s all it takes.
2. Make it safe to experiment. Create space for people to try things and fail. Many people are afraid to use AI at work because they’re not sure what’s “allowed” or what counts as good output.
3. Share real examples internally. When someone on your team uses AI to solve a problem or speed up their work, share it. Make it visible. Make it normal.
Sit down with your CFO. Your head of CS. Even Your VP Product. Show them what’s possible.
We talk about AI like it’s everywhere. Like everyone’s using it. Like we’re in the late innings.
We’re not.
We’re in the second inning. Maybe the third.
The companies that win the next few years won’t just be the ones with the best AI strategy. They’ll be the ones who actually got their teams using these tools day-to-day.
So take an hour this week. Sit down with your CFO. Your head of CS. Your VP Product. Show them what’s possible.
You might be surprised how many minds get blown.
And how many weekly update emails you start getting about it.
