The planning template for next year arrived this week. It looks almost exactly like last year's, with the same tabs for headcount and spend, and one new tab at the end called "AI Initiatives." It has four columns, running from initiative to timeline. It is due a week from Friday.
Somebody in finance built that tab in an afternoon, and whatever you type into it will follow you around for twelve months.
Here is my opinion about that row, and I hold it fairly strongly. Anything you write in a planning document becomes a promise, and planning promises get checked in December by people who were not in the room when you made them. So I'd write the row you would be happy to be held to. That is rarely the most impressive one on the page, and it is almost always the one that ages best.
What the tab is really asking
The template says "initiatives," but three different people will read your answer, and each of them wants something different.
Your boss wants something they can repeat upward without being embarrassed. Finance wants to know whether to hold money for you and whether any of your costs are about to go down. Your boss's boss will skim the tab for the word AI, feel reassured that it appears, and move on.
None of them is asking for a vision. Nobody has ever been promoted for the adjectives in a planning template. What survives the year is a row that is small, specific, and true, because in December it will still say what you did.
The four lines I'd write
I'd fill in one initiative, not five, and I'd make each column do one honest job.
The initiative is a named task, not a theme. "Use AI to improve customer experience" is a theme, and a theme can never be finished, which means it can never be done well either. "Draft the weekly customer status report with the tool the company already allows, with a person checking every figure" is a task. You can tell in March whether it happened.
The benefit is in hours, with its source. If you have already tried the tool on that task, write what you saw and say so: "Estimated two hours a week, based on a four-week test in October." If you have not tried it yet, say that instead: "To be measured in the first quarter, readout by the end of February." An honest "to be measured" is worth more than a confident number you made up on a Thursday afternoon, and finance can smell the difference.
The cost has a ceiling. Seats, times the price per month, times the number of months, written as a total. Add the words "no new headcount" if that is true. A cost with a ceiling is a decision. A cost without one is a subscription that renews itself.
The timeline is a decision date, not a launch date. "Expand to one more task or stop, by the end of March, based on the readout." That line does more for you than anything else in the row, because it shows you planned to find out whether it worked.
Here is the row as a template. Copy it into the tab and fill in the brackets.
Initiative: [one named task] for [team], using [the tool your company already allows].
Expected benefit: Estimated [hours] per [week or month], based on [the test you ran, or "a four-week measurement in the first quarter, readout by (date)"].
Cost: [seats] × [price per month] × [months] = [total]. No new headcount.
Decision point: Expand to one more task, or stop, by [date], based on the readout.
Four lines. No vendor names in the benefit, no adjectives anywhere, and a date on which you will know more than you do today.
The one number I would leave out
The column you will be most tempted to fill in is the one nobody put on the template: headcount.
Somewhere above you, someone would love to see a line that says the tool saves one person. I would not write it for a tool you have not measured, and here is why. Once a headcount saving appears in a plan, it tends to come out of your budget in January, whether or not the tool delivers by June. Hours saved go back into the team. A headcount saving comes out of it, and it comes out on the planning calendar, not the tool's.
So I'd describe the gain in hours and say what the team will do with them. Name the thing that has sat on your backlog for a year because nobody had time: the process document, the customer follow-up, the cleanup nobody owns. "The saved time goes to the renewal follow-ups we currently skip" is a sentence finance can live with, and it keeps the benefit inside your team where it belongs.
If someone senior insists on a headcount figure anyway, I'd say this, and you can use it as written:
"I'd rather that number come from you, because I can't defend it yet. What I can defend is the hours, and I'll have measured ones by March."
It is polite, it is true, and it puts the number with the person who wants it.
The note to send with it
Do not let the row travel alone. Send your boss a short note when you submit the template, so they hear your reasoning from you before they read it in a spreadsheet. Something like this:
Subject: The AI line in our plan
I've kept our AI line small on purpose. It's one task we can measure, with a cost ceiling and a decision date, and I'd rather beat it than explain it next December. If the readout is good in March, I'll come back with the next task and a number I can stand behind. Happy to talk it through before it goes up.
Four sentences. It tells them you were careful rather than timid, and it gives them a line to use if their own boss asks why your row looks modest.
If your boss wants it bigger
Sometimes the answer comes back, "Can we be more ambitious here?" That usually means someone above them is counting rows.
I'd agree to more ambition and keep the shape. Add a second named task with its own decision date, rather than inflating the first one. And say this:
"I can make it bigger. I'd just rather make it bigger in March with our own numbers than now with a vendor's."
Most bosses hear that as good judgment, because it is. The ones who still want a bigger promise are telling you something useful about how the year is going to go, and it is better to learn that in October.
What this gets you
When the template goes in, your AI row will be the shortest one on the tab. It will name one task, give one honest estimate, cap one cost, and set one date.
In December, when finance opens the plan and runs down the rows, most of that tab will read like a brochure from a year ago. Yours will say what you did, what it cost, and what you learned. That is rare enough in planning documents that people notice, and the people who notice tend to be the ones who decide who gets next year's interesting work.
Fill in the row, send the note, and close the template. It will still be there on Monday. Go enjoy the rest of your Thursday.
P.S.
If you want to know where your team really stands before you write the row, The AI Readiness Self-Audit is a twenty-minute snapshot across five areas, scored by you, on one page. ($19. It makes the "expected benefit" column much easier to fill in honestly.)
This was a free one. There's a new one every week.
What's signal in the AI noise — and the move to make about it. No hype, no vendor pitch, no link dump.
Aiden Vector is an AI-assisted publication; this content is produced by AI under human editorial direction.