How executives should learn AI in 30 days: a hands-on plan built on your own calendar
Most AI training for executives is a slide deck about the future. You do not need the future explained; you need to know what this does for your Tuesday. The fastest way I know to get there is thirty days of using it on your own work, with a little structure. Here is the plan I give the operators I coach.
the short version
- Learn on your own work, not on toy examples. Your inbox, your board memo, and your weekly numbers are the curriculum.
- Spend about thirty minutes a day. Consistency matters more than intensity.
- By day thirty you should have three tasks you no longer do by hand, one you deliberately kept, and a clear view of where AI fits in your company.
- In my experience, executives who use AI personally make better calls about where to invest in it, and their teams adopt faster when they see the boss using it.
00before day one
Three things to set up, which should take less than an hour:
- An approved tool. One business-grade assistant your company has cleared for company data. If you are choosing between the two most common ones, read Claude vs ChatGPT for business, then pick one and commit for the month.
- A time slot. Thirty minutes, same time every day. First thing in the morning works for most people because the work is fresh.
- A running log. A single document where you write what you tried, what worked, and what did not. Two lines a day. By day thirty this log is the most valuable output of the month.
01week one: reading and summarizing (days 1 to 7)
Start with the lowest risk, highest frequency work you do: reading. Executives spend an enormous amount of time getting through material to find the parts that matter.
- Day 1: take the longest document in your inbox and ask for a summary, then ask what is missing, what is unclear, and what the author is asking you to decide.
- Day 2: give it a contract or proposal you already know well. Check its summary against what you know. This is how you calibrate trust.
- Day 3: paste in last week's meeting notes and ask for decisions made, open questions, and owners.
- Day 4: ask it to read two competing proposals and lay out the differences in a table.
- Day 5: ask it to argue against a decision you are leaning toward. Notice whether the objections are new to you.
- Days 6 and 7: repeat whichever of these saved the most time, on real work.
The lesson of week one is that the tool is fast and usually right, and occasionally wrong with total confidence. The term for that is hallucination. Day 2 is where you learn to spot it.
02week two: drafting (days 8 to 14)
Now move from reading to writing. The goal is never to send something you did not read. The goal is to stop starting from a blank page.
- Day 8: write a short guide to your own voice: three emails you have sent that sound like you, and a few rules ("short sentences, no jargon, end with the ask"). Save it and reuse it.
- Day 9: draft five routine replies with it. Edit them. Note how much editing each one needed.
- Day 10: turn rough bullet notes into a memo for your leadership team.
- Day 11: prepare a brief for tomorrow's most important meeting: who is attending, what they want, what you want, likely objections.
- Day 12: draft a difficult message, such as a no to a partner or feedback to a direct report, then rewrite it yourself with the draft as a reference.
- Days 13 and 14: review your log. Which drafts needed light edits and which needed rewrites? That tells you what to delegate.
03week three: numbers and decisions (days 15 to 21)
This is the week most executives find surprising. Modern assistants can work with spreadsheets and data exports, not just text.
- Day 15: upload an export of last month's sales or operating numbers, with anything sensitive removed if your tool is not approved for it, and ask what changed and why it might have changed.
- Day 16: ask it to build the chart you always ask your analyst for. Compare the result.
- Day 17: describe a decision you need to make and ask for a simple decision framework: options, criteria, what you would need to believe for each option to be right.
- Day 18: ask it to stress test a forecast: what assumptions carry the most weight, and what happens if each is wrong.
- Days 19 to 21: use it in a live working session with one direct report on a real problem. Watching someone else use it changes how you think about rolling it out.
Always check numbers that matter against the source. AI is good at structure and narrative around data. It can still misread a column. The habit of checking is part of the skill.
04week four: your company, not just your job (days 22 to 30)
By now you know what the tools can do for one person. The last stretch turns that into judgment about your company.
- Day 22: list the ten most repetitive tasks in your company that you know of. Ask your direct reports to add theirs.
- Day 23: for each, note the volume, who does it, and what a mistake costs. That is a rough opportunity list.
- Day 24: learn the difference between a chat assistant and an AI agent that takes actions in your systems. It changes what is possible, and what needs controls.
- Day 25: ask your teams what they already use. Some of them are using personal accounts with company data. Better to know.
- Days 26 to 28: pick the top two opportunities and write one page on each: the task, the volume, the current cost, and what good looks like.
- Days 29 and 30: review your log and write your own policy for your own work: what you now always delegate to AI, what you never do, and what you are still testing.
05what to delegate, what to keep, what to stop
| Delegate to AI | Keep for yourself | Stop doing |
|---|---|---|
| First reads of long documents | Final decisions | Reading entire reports to find one section |
| First drafts of routine messages | Hard conversations | Writing status updates from scratch |
| Meeting prep briefs | Relationship judgment | Asking for decks that only summarize |
| Arguing the other side | Setting strategy | Waiting days for simple analysis |
06how to measure the month
You are not measuring company ROI yet. You are measuring whether the habit stuck and whether your judgment improved.
- Hours returned per week. Estimate from your log. Be honest, including the time spent fixing bad output.
- Tasks moved. How many recurring tasks now start with AI.
- Quality of the opportunity list. Do you have two written opportunities with volume and cost attached? That is the bridge to company-level investment.
- Team pull. Are your direct reports asking how you did something? Adoption follows the boss.
07doing it with a coach
You can run this plan alone. Most executives I work with go faster with someone who reviews the log each week, unblocks the tasks that stall, and connects what you learn to decisions your company is facing. That is what AI coaching for executives is: this plan, applied to your actual calendar, one on one with me.
When the month ends and the question becomes "how do we get the whole team here", that is AI training for teams. And when the opportunity list turns into something worth building, that is AI implementation, scoped against the same numbers you wrote down on day 23. I would rather you arrive at that conversation with your own judgment than with someone else's slide deck.
common questions
How long does it take an executive to learn AI?
A working command of AI tools for your own job takes about a month of daily use, roughly thirty minutes a day. Understanding where AI fits across your company takes longer, but it builds on the same habit, because your judgment improves with every task you try yourself.
Which AI tool should executives start with?
Start with one general purpose assistant your company has approved, such as Claude or ChatGPT on a business plan, so your data is handled under business terms. Pick one and stay with it for the month. Switching tools every few days resets the learning.
Is it safe for executives to put company data into AI tools?
Only in tools your company has approved under business terms that cover data handling and retention. Personal consumer accounts are the wrong place for board materials, financials, or customer data. Ask your technology lead which tool is approved before day one.
What is the difference between AI training and AI coaching for executives?
Training covers how the tools work, usually in a group. Coaching is one on one and applies the tools to your specific job, calendar, and decisions, with someone reviewing what you tried and adjusting the next step. Most executives need a little of the first and more of the second.
What should an executive delegate to AI first?
First drafts and first reads: summarizing long documents, drafting routine replies, preparing meeting briefs, and turning rough notes into structured memos. You still review and decide. The savings come from never starting from a blank page or reading a hundred pages to find three that matter.