Your roadmap is a list of guesses wearing a suit

Here’s a thing nobody tells you when you get handed your first roadmap: it’s fiction. Well-dressed fiction. It has quarters and swimlanes and little rounded rectangles in three shades of blue. It looks like a plan. It behaves like a promise. But strip the formatting away and every box on it is the same thing — a guess about the future, typed with confidence, in a font that says trust me.

I don’t mean that cynically. I’ve made these documents. I’ve defended them in rooms. The suit isn’t a lie you tell other people so much as one you tell yourself, because a spreadsheet of guesses is unsettling and a roadmap is calming, and calm is what gets you out of the meeting.

Where the suit comes from

A guess becomes a roadmap through a specific laundering process. First you estimate — “this feels like about six weeks.” Then someone asks for a date, so six weeks becomes March 14th. Then the date goes in a slide. Then the slide goes to a customer. Somewhere in that chain the feels like quietly evaporates, and what’s left is a number with no memory of how soft it used to be.

The tell is that nobody writes down what the guess was resting on. “March 14th, assuming the API team ships auth on time, assuming the design doesn’t change, assuming nobody quits, assuming this is even the right feature.” Those clauses are the actual content. The date is just their shadow. But we ship the shadow and delete the clauses, then act surprised when reality audits us.

A roadmap doesn’t fail because the guesses were bad. It fails because you dressed them so well you forgot they were guesses.

There’s a real idea underneath this worth knowing by name: the cone of uncertainty. Early in a project your estimate can be off by a factor of several in either direction, and that range only tightens as you actually build and learn. The unnerving part is that the cone doesn’t narrow because time passes. It narrows because you do the work. Sit on a project for a month doing nothing new and you’ve learned almost nothing; the uncertainty gets burned down by building, not by planning harder. Which means the most detailed roadmap — the one drawn at the exact moment you know the least — is precisely the one making the boldest claim.

What to do instead

You can’t stop guessing. Guessing about the future is the whole job. So don’t try to make better roadmaps; try to make more honest ones.

Three moves that have actually helped me:

Date the guess, not the delivery. Instead of “ships March 14,” write “as of January, we believe roughly late Q1.” Same information, work clothes instead of a suit. It invites revision. It survives contact with reality because it never claimed to have escaped it.

Attach the assumption to the box. Every item gets one line: true only if ___. When that assumption breaks — and one always does — you have a paper trail explaining why the date moved, instead of a mystery that reads as failure.

Separate what you’ve committed from what you’re exploring. Most roadmaps blend a few real promises with a pile of maybes and render them identically. Give them different weight. Near-term, high-confidence work can look solid. Anything past the next horizon should look sketchy — literally, dashed and faint — because it is.

The goal isn’t to look less competent. It’s the opposite. The people I trust most with a plan are the ones who can tell me, precisely, which parts they’re sure of and which parts they’re bluffing. That’s not weakness. That’s the highest-resolution picture of the future anyone can honestly hand you.

So keep the roadmap. Just take its suit off now and then and remember what’s underneath: a row of guesses, doing their best, waiting to be settled by the only thing that ever settles the matter — the build itself.