There's something a trainer probably shouldn't say in public, so naturally I'm going to say it: the most commercially successful format I deliver is also the one I trust least to change how anyone actually works. The one-day AI intensive sells because it's administratively frictionless. Easy for a manager to approve, one line in a budget, no recurring commitment. Easy to schedule, one day out of the office rather than a drip of interruptions across a quarter. Easy to report, a tidy row on the training dashboard, headcount trained, box ticked. Every one of those advantages is real. And not one of them has anything to do with whether behaviour changes.
The learning research has been saying the same thing for decades, and my own experience confirms it with depressing regularity: durable skill comes from spaced practice with feedback, not a single intense exposure. Eight hours of new material, however well delivered, mostly decays within weeks if nothing in the participant's actual work reinforces it. I've had participants greet me warmly months after a session and, in the same friendly conversation, reveal they'd retained roughly two things, usually the funniest demo and one tool name. These were engaged, intelligent people. The format simply did what the format does, and I'd sold it to them anyway.
I still run one-day workshops, and I'd rather be honest than pious about why. They pay the bills, clients ask for them, and they're a legitimate low-commitment way for an organisation to test whether AI training is worth taking seriously. I'm not above the economics of my own industry, and anyone in it who claims to be is selling you something else. What I have stopped doing is pretending, to clients or to myself, that a single day rewires working habits. For a while I did pretend, or at least allowed the pretence to stand unchallenged in scoping calls, because correcting a client's expectations mid-pitch is uncomfortable and I was newer to this and wanted the booking. That was a disservice to everyone, including the format itself, which fails when measured against "transformation" but does fine against what it can actually deliver.
So here's the repositioning that let me sleep better. I now treat one-day sessions explicitly as diagnostics. A day in a room with a team tells me what they genuinely don't know, as opposed to what their L&D lead assumed they didn't know, which is reliably a different list. It surfaces their real workflows, their shadow AI habits, their sceptics, and the one or two people who might become internal champions. All of that is scoping gold for designing something longer, spaced out, and attached to their actual work, which is where behaviour change actually lives. The day becomes the intake assessment rather than the treatment. Clients who hear it framed that way almost always prefer it, which suggests the pretence was never necessary in the first place. I could have skipped a few years of it.
Exhibit · The repositioning
The one-day workshop, reframed
Sold as
Transformation
- A single day rewires working habits
- Administratively frictionless to approve
Actually is
A diagnostic
- Surfaces real workflows and shadow AI habits
- Identifies sceptics and potential internal champions
- Scopes what to build next
If you're an L&D lead whose entire AI training budget is one-off workshops, I'd gently suggest you're looking at a format problem dressed up as a training decision. The question worth putting to any provider, very much including me, isn't what the day covers. It's what happens in week three. If the answer is silence, you're buying a pleasant day and a dashboard entry. Sometimes that's honestly all that's needed. But it's worth knowing that's what you bought.
