When you hire an AI training consultant, three things predict whether the training works: real teaching ability, a curriculum built for your team instead of a template, and a habit of understanding your people before proposing content. A polished demo tests none of them. This post covers what to check and what to ask before you sign.

The market got crowded fast. Plenty of people who did something else 18 months ago now have "AI trainer" in their bio. Some are excellent. Some know the tools but have never taught anyone. Telling them apart before a contract is the whole job.

The real problem: knowing AI is not the same as teaching AI

Demand is real and mostly unmet. Pew Research Center found that among workers who took any job training in the past year, only about a quarter said it touched AI, and roughly half of workers who already use AI said their employer offered no training on it.

That gap pulls in consultants whose main qualification is that they use AI a lot. Being fluent with a tool and being able to teach it to a mixed-skill, non-technical team are two different skills. One is expertise. The other is instructional design. You're hiring for the second one.

1. Instructional design experience, not just AI knowledge

This is the filter that matters most, and it's the easiest to skip because AI knowledge is louder in a sales call.

Anyone can walk a room through a tool they like. Far fewer can sequence a topic so a nervous beginner and a confident power user both leave better off, notice the person who's quietly lost, and design practice that makes the skill hold up after everyone logs off.

The stakes are measurable. Research on training transfer consistently finds that only a small share of what's taught gets used on the job, often cited around 10 to 12%, decaying further over the following months without reinforcement. Instructional design is the discipline built to close that gap. A consultant who's never studied how adults learn will likely deliver a session people enjoy and then forget.

Ask directly: how do you design a session for mixed skill levels, and what do you build in so it holds up afterward? Listen for a method, not enthusiasm.

2. Custom-built curriculum, not off-the-shelf slide decks

The tell is simple. Does the consultant ask about your workflows before proposing content, or pitch you the same deck they pitch everyone?

Off-the-shelf training is cheaper to deliver and easy to sit through, but the examples aren't your examples, so the jump from "interesting" to "I'll do this Monday" never happens. Custom-built training starts from your team's real tasks, so the practice is the work.

You can spot this on the first call. A consultant quoting a fixed curriculum before understanding what your team does is selling a slide deck. A consultant asking questions before proposing anything has the right instinct.

3. Meeting your team at its actual skill level

A good consultant is comfortable whether your team is brand new to AI or already using it inconsistently across the group.

That second case is more common than people admit, and trickier. When half the team is ahead and half is anxious, generic "intro to AI" content bores one group and loses the other. The right consultant checks where people are before deciding what to teach, rather than assuming a starting point.

Questions to ask before you sign

Take these to any consultant you're seriously considering. The quality of the answers tells you more than the answers themselves.

  1. What will you do to understand our team before you build anything? Listen for a real discovery step, not "I'll send the standard materials."
  2. How do you handle a room with mixed skill levels? Listen for a method.
  3. Is your curriculum custom or off-the-shelf, and how much gets tailored to us? Off-the-shelf isn't automatically wrong, but they should say so plainly.
  4. How will we know it worked? Listen for something past attendance. Behavior change and follow-up beat a happy-sheet survey.
  5. What's your background in teaching or instructional design, not just using AI? Listen for evidence they've taught before.
  6. What happens after the session? Listen for reinforcement. One-and-done rarely holds.

A consultant who gets defensive at these questions has given you your answer.

What a gap-analysis-first engagement looks like

Here's the approach I take, described plainly so you have something to compare against. (This is my process and my professional opinion about what works, not a standard every consultant follows.)

I don't start with content. I start with a gap analysis: where the team is now, where it needs to be, and the real distance between those two points. Design only makes sense after that, because the design should close that specific gap.

In practice, the first conversation is mostly me asking questions. What does the team do day to day? Where does work get slow or error-prone? Who's ahead, who's anxious, what have they already tried? What constraints, like budget, time, or compliance, are real? The program gets built from those answers.

Content first, team second is how you get training that demos well and changes nothing. I'd rather spend the first hour understanding the gap than the first session filling the wrong one.

The short version

Hiring an AI training consultant well comes down to three checks: real teaching ability, a program built for your team, and a habit of learning where your people are before starting. A demo answers none of those. The questions above do.

This is the deep-dive companion to our homepage FAQ, What should I look for when hiring an AI training consultant? If you want to see what a gap-analysis-first conversation feels like, that's what the free 15-minute call is: no deck, no pitch, just questions about your team and an honest read on what would help.

Sources

Notes marked as my process or professional opinion are exactly that, not external data. Transfer-of-training figures come from research summaries, vary by study, and describe training generally, not AI training specifically.