Who to Train First When You Roll Out AI Literacy
By Mykel Stanley, StrategixAI
The first question a CTO asks me when we sit down to plan an AI literacy rollout is usually the wrong one. It is some version of "how long is the course." The right question is who goes through it first.
Sequencing matters more than curriculum. I have seen well-designed training programs die because the wrong cohort went through them first, and I have seen mediocre training change a company because leadership got it right with the first thirty people. If you are spending real money on AI literacy training, the order you train people in is the single biggest lever on whether the investment shows up on the P&L.
Why the First Cohort Decides Everything
Mid-market operations are political. A new initiative gets evaluated in the hallway during the first ninety days, not in the survey at the end. If the first wave through your AI literacy program walks out skeptical or unconvinced, every future cohort is harder to enroll. If they walk out applying what they learned the next morning, your second cohort doubles in size.
This is not a curriculum problem. It is a sequencing problem. The people who go first set the narrative for everyone who comes next.
The Three Groups That Should Go First
In almost every mid-market rollout I have run, the right order is the same.
First: the operations leaders who own a P&L line. Plant managers, distribution center directors, controllers, VPs of operations. These are people who can decide on Monday morning to change a workflow and have the authority to make it stick. When they finish a literacy program with a clear next step in their own operation, the program becomes real for everyone reporting to them. Skip this group and you are training people who cannot act on what they learned.
Second: the team leads and supervisors in the workflows you actually want to change. Once the P&L owner has a direction, the next layer translates it. The shift supervisor in a manufacturing line. The senior estimator at an engineering firm. The dispatch lead at a logistics company. These are the people who will be evaluating AI output every day. If they cannot read what an agent is doing, the agent does not get used or it gets used badly.
Third: the IT and security leads who will set the guardrails. Notice this is not first. The most common mistake mid-market companies make is starting AI literacy with IT, because IT is who asked. That trains the people who can say no before training the people who can say yes. IT literacy is essential, but it lands better after the business side has a real use case to govern.
Who Should Wait
The marketing manager who is excited about ChatGPT does not need to be in the first cohort. Neither does the analyst who has been posting on LinkedIn about prompt engineering. They will get there in cohort three or four, and they will get more out of it when the people above them already know how to evaluate their work.
Same logic for the executive who insists on being in the first session because they want to set the tone. Put them in their own short briefing. The first real cohort should be people who will go back to a team and change how that team works the following week.
What Goes Wrong When the Order Is Off
I have walked into companies six months after they ran AI literacy training and asked what changed. The answer is usually nothing. Then I ask who went first. The answer is almost always one of two things. They trained IT and security first, and the business never built a real use case to govern. Or they trained whoever raised their hand, which usually means individual contributors with no authority to redesign a workflow.
Both versions produce the same outcome. People know more about AI. The operation does not change. The CFO concludes training did not work and pulls the budget. It worked. The sequencing did not.
What This Looks Like in Practice
A 400-person manufacturing client of ours sequenced their first three cohorts this way. Cohort one was the plant manager, the production manager, the quality director, the controller, and the VP of operations. Twelve people. They left with three workflow changes already scoped. Cohort two was twenty-two team leads across two shifts and quality. By the end of cohort two, the AI-assisted quality inspection workflow we had scoped in cohort one was running on the line in pilot. Cohort three was IT, security, and HR. By then the program had momentum, and IT was governing something real instead of theorizing about it.
That is what good sequencing looks like. Decisions, then translation, then governance. Not the other way around.
How to Pick Your First Cohort This Quarter
Look at your org chart. Identify the five to twelve people who own outcomes you want AI to influence in the next two quarters. Not job titles, outcomes. If you cannot name a workflow they would change on Monday morning after training, they are in the wrong cohort.
Then build the rest of the rollout around them. Their direct reports go next. Their peers in adjacent functions go in cohort three. IT, security, and HR come in to govern what is already moving.
At StrategixAI, this is how we structure the AI Literacy Pipeline for mid-market clients. The training is only part of the engagement. The sequencing is what makes the training actually change the operation.
If you are about to spend real money on AI literacy and you have not decided who goes first, book a consultation and we will walk through your org chart together. Get the first cohort right and the rest of the program runs itself.