A list tells you what’s outstanding. A plan tells you where you are in the year and what you’re allowed to say no to. Those aren’t the same document, and treating them as the same document is how a well-intentioned PMM ends up with forty open items and zero sense of whether this is a good week or a bad one.

You Have Too Many Calendars

This problem lands harder on PMM than most functions because everyone else’s calendar is legitimate and nobody’s job is to reconcile it with anyone else’s: Product runs a shifting sprint-by-sprint roadmap, Sales runs an annualized quota calendar plus whatever’s closing Thursday, Marketing runs a campaign calendar built months out, Events run on a conference organizer’s T-minus workback. PMM sits in the middle of that Venn diagram, which means a plan that only covers “my department’s stuff” doesn’t survive first contact with reality — it needs to work at more than one zoom level at once.

Three Altitudes

The Annual View. The map, not the itinerary — what’s already fixed on the calendar this year: major launches, seasonal windows, roadmap milestones firm enough to plan around. Keep it short; more than eight or ten entries and it’s stopped being a map and started being a list wearing a map’s clothing.

The Quarterly Forecast. Where the annual view gets honest. A year-out roadmap is a hypothesis; this quarter is closer to a commitment. This is the altitude where trade-offs get named out loud — if a launch slips from Q2 to Q3, that’s a conversation about what else in Q3 now has to move. Reforecast every quarter on purpose, not just when something breaks.

The Weekly List. The ground floor — the only altitude most PMMs have ever built explicitly. Necessary but nowhere near sufficient. The test for anything landing here: does it trace back up to the quarterly forecast? If not, it’s either a genuine opportunistic item or a random act of marketing wearing a task’s mask (see Avoiding Random Acts of Marketing).

The 80% Rule

Plan to no more than 80% of capacity — not as a wellness platitude, but structurally. You already know, with near certainty, that opportunistic work is going to land on you. Not which piece, not exactly when, but that something will. A plan built at 100% capacity is wrong on the day you finish writing it, because it has nowhere to put the thing you knew was coming. Twenty percent headroom is the one honest way to plan around a category of work you can predict the volume of but not the specifics.

Where AI Helps

Not primarily by drafting faster — by reconciliation. The three altitudes only stay useful if they stay in sync with each other, and keeping three moving documents in sync by hand is exactly the kind of unglamorous bookkeeping that quietly stops happening a few weeks after the last scrub. Feed AI your weekly list and quarterly forecast and have it flag the moment a slipped weekly task means the quarter needs updating, or the moment a quarter is drifting from the annual view before it surfaces secondhand in a leadership meeting. It can’t decide what the plan should say. It can make sure the plan still tells the truth about itself, at every altitude, every week.

Plan on a Page

All of this collapses into one page built for a VP review: role and scope, big moments and milestones laid out as an actual timeline, FY focus (how it ladders to the business), measures of success, needs for success (budget, capacity, dependencies), and ownership and cadence. Fill it in as a team, and treat it as something meant to be argued with, not admired.

Full post and a fillable template you can edit and export as an image, Markdown, or PDF: PMM Plan on a Page →