Last week I asked you to look at week four instead of launch day. If you did, you probably saw the slide. This week is about the first place leaders reach when they see it, and why it costs the most.
The slide shows up in month two. Somebody in the meeting says the platform's clunky. Too many clicks. The vendor oversold it. A committee forms, demos get booked, and eighteen months later there's a new platform, a new kickoff and a new dashboard.
And week four arrives again, on schedule, with the same notebook on the same desk.
I've watched companies do this three times in a row with three different vendors and the same result each time.
Swapping the tool doesn't fix week four. It resets the clock so that week four looks like launch day for a while. Every swap moves the money without moving the outcome.
Find two teams that got the same tool, the same week, from the same trainer. Look at week four for each.
In 29 years I've never seen the two lines match. One team is living in the new system: records landing from the parking lot, managers coaching off a dashboard that's true. The other is keying it in at six o'clock from a notebook.
Same software. Same license. Same deck. If the tool were the problem, both lines would slide together. They never do. So the variable is inside the building, and it doesn't appear on a feature comparison.
BCG's long-standing rule of thumb for AI that actually works in a company is roughly ten percent algorithms, twenty percent technology and seventy percent people and process. Writer's 2026 enterprise survey found 79 percent of executives reporting adoption challenges, a double-digit jump in a single year, during the best year the software itself has ever had. The tools got better and the slide got worse.
When the tool improves every quarter and the outcome doesn't move, the tool isn't the lever. Seventy percent of it never was.
Two floors of the same building, same platform, same launch week. On four, week four looked like the demo. On six, it looked like every rollout you've ever seen.
The VP on six was certain it was the tool. Clunky, he said. So I asked the manager on four how many clicks it took her people. "Same as six," she said. "I just stopped letting the old way be an option in week two."
One sentence. Same software. He'd spent a month reading vendor reviews.
When the slide shows up, leaders split by the first question they ask. Group one asks which vendor they should have picked and spends the next year in demos. Group two asks which team held on and what their manager did, and spends the next week copying one conversation.
The second group isn't more technical. Half of them can't tell you what the platform does. They just refuse to believe the answer is in a feature list when the evidence is on the floor.
Before anyone books a demo, find your floor four. Somewhere in the company the same tool is holding in week four. Go ask that manager what she did in week two. Write it down. That sentence is worth more than the next platform.
Next week I'll take apart the second reflex, more training, and show you why the drop starts on the day the trainer leaves.
If you want the number for your own team, the Caveman Code Diagnostic takes two minutes.
Dr. Noah St. John, The Caveman Conversion King
Founder of NoahMentor.com
