Jared Lukes. Start here

AI and your people · Updated 2026-10-19

Can't I just do the AI stuff myself?

Probably, yes. Most owners I meet are already good with the tools. What they cannot do is get the rest of the team to use them the same way, or keep it going on the days they are out. That is not an AI problem, it is a teaching problem, and it is cheaper to fix than people think. The first step is $450.

This is the question behind most of the quiet "no thanks" emails I get. People hate paying for AI help because they figure they can do it themselves. A lot of the time they are right.

Can I do it myself?

If you have spent a few evenings with ChatGPT or Copilot on real work, you are ahead of most consultants who will try to sell you a strategy. You do not need me to tell you what a prompt is.

Where it breaks is the next step. You know how to get a good answer out of the tool. Your bookkeeper does not. Your front desk does not. Your best salesperson is doing it their own way and nobody knows what way that is. The business is only as good at AI as the person who is out sick that day.

Why can't I just show them?

You can, and you should. Three things usually get in the way.

  • You are busy. Showing someone properly takes a morning, and the morning never comes.
  • You do it by feel. You cannot explain what you do because you never had to. Teaching it means writing it down, and that is a different skill from doing it.
  • They are scared of looking slow. People will not ask the owner a dumb question. They will ask me.

Putting what you already know into the hands of the people who work for you is the whole job. That is what "putting a robot suit on your employees" means in practice. The suit does not work if only the boss knows how to wear it.

What does the cheap version look like?

Ninety minutes at your place, $450. I turn on the AI that is already inside the tools you pay for, with the team watching, on their actual work. Your email, your spreadsheets, your accounting software. Most businesses have never switched half of it on. Everybody leaves with a one-page sheet: what to use it for, what never to put into it, and who to ask.

That is usually enough for a business under ten people. If it is not, the full day with the work on the wall is $3,500, and the $450 comes off it.

When should I not hire anyone?

When nobody on the team is curious yet. A session for people who do not want to be there is wasted money. Spend the $450 on lunch instead, and let the one curious person on staff show the others what they found. Call me when there are two of them.

What does it cost to get it wrong?

Usually nothing dramatic. The tool sits there unused, you keep paying for it, and the person who was good at it leaves and takes the knack with them. The quiet cost is the hours that never came back.

Answered by Jared Lukes, Minneapolis. Twenty-five years in brand, UX, and technology, currently holding fractional AI and technology seats at three companies. About Jared.

Quick answers

Is $450 really enough to be useful?
For a team under ten, yes. The tools you already pay for are most of the first win, and switching them on properly is a morning's work.
What if I am the only one who uses computers much?
Then the session is for you, and the one-page sheet is for everyone else. We write it so a person who hates software can follow it.
Does the $450 session work over a screen share?
Yes. On site is better for the first one, because people ask more questions in a room.
Will you try to sell me a bigger thing afterward?
I will tell you whether a bigger thing is worth it. If it is not, I will say so and you will not hear from me again unless you call.
Can you train the one person who then trains everyone else?
Yes, and that is often the best money. Give me your most curious employee for an afternoon and you get a trainer who works there.

Got a version of this question that is more specific?

A 30-minute call, no deck, no pitch. Tell me what is stuck and I will tell you straight if I can move it.