Week one produces screenshots. People log in, try a prompt and forward the result to the sponsor. The following Wednesday the old pack is back, built the way it was built in September.
The licence is live. The meeting still asks for the old artefact, on the old deadline, with the old definition of done. Anyone who used the tool has to do the work twice, once to learn and once to satisfy the review. By the second week the second pass wins. The tab stays closed.
Use is common and the workflow is old
McKinsey's 2025 State of AI survey found that 88 percent of respondents report regular AI use in at least one business function. About one-third say their companies have begun to scale. The majority are still experimenting or piloting. The same survey says most organisations have yet to embed the tools deeply enough into workflows and processes to see material benefits across the enterprise.
That gap is the week-two stall in practical form. A login counts as use. A workflow still expects the old output, so people serve the meeting. When the reviewer still marks up last quarter's template, the new tool loses before the next Monday. The dashboard records a launch. The work records a return to habit.
McKinsey also reports that the companies seeing the most value are nearly three times as likely as others to say they have fundamentally redesigned individual workflows, and that this redesign is one of the strongest contributors to business impact in the survey. A rollout that leaves the workflow untouched will collect activity in week one and a quiet screen after that.
What AI adoption training has to change
AI adoption training should pick one task a named person already owes, and change the way that task is produced and reviewed. The prompt, the source material they are allowed to use, and the check a human still does all have to be written down. The review meeting has to accept the new output. If the meeting still wants the old pack, the session was a tour.
Our AI training workshop is built for that task. Half a day or a day, on the team's real work, with guardrails for data and accuracy. Everyone leaves with a rep they can run the next morning, and with the reviewer in the room so the check is visible. What a single day should produce is in what a corporate AI training day should produce. What I want a manager to change with their own team is in AI training for managers.
The second Wednesday
Before you book, name the task and the meeting that will consume it. Name the person who reviews the work, and put them in the room. Agree what done means once the tool is in the path. If you cannot name those three, the rollout will look busy for a week and familiar after that.
Send me the task, the reviewer and the meeting. We will shape the day around that work. A reviewer who cannot attend is a reason to move the date. The second Wednesday is the test. If the new output is what the meeting accepts, the rollout has started. If the old pack is back, you bought a login.
