You shipped on time. Nothing changed.
There’s a particular meeting that happens in every growing software company. The quarter closes. The team shipped everything on the roadmap — on time, tested, demo went well. And somewhere in the metrics review, someone notices that the numbers didn’t move. Activation is flat. Retention is flat. Revenue did what it was going to do anyway.
Nobody says anything sharp, because what is there to say? Everyone did their job. The engineers built what was asked. The designers designed what was specced. The PM delivered the plan. The plan was the problem, and the plan has no one’s name on it.
Output is what you did. Outcome is what changed.
The distinction sounds like consultant vocabulary until you price it.
Take a team of thirty engineers. Fully loaded, that’s roughly ninety engineer-months of capacity a quarter. Now run the honest audit: of the things you shipped in the last two quarters, how many moved a metric anyone would defend in a board meeting? In most companies we look at, the honest answer sits between a third and a half. Call it conservatively 30% pointed at things nobody needed — that’s 27 engineer-months a quarter, spent producing software that works perfectly and does nothing.
No engineering improvement touches this number. Ship twice as fast and you produce twice as much of it.
Why good teams end up here
Not incompetence. Structure. Three mechanisms do most of the damage:
The roadmap is a treaty, not a strategy. Sales needs the enterprise feature to close the deal. The biggest customer is threatening to churn without their integration. The CEO promised something on a podcast. Each item is individually defensible; collectively they’re a list of whoever asked loudest, and lists of requests don’t compound into anything.
Shipping is measurable and outcomes are slow. A release happens on a date; retention moves over quarters. Organisations manage what they can see weekly, so velocity becomes the score. Teams get very good at the thing being scored. That’s not a pathology — it’s people responding rationally to the scoreboard they were given.
Nobody owns the question “why this?” Engineers own “how.” PMs, in most companies, own “when.” The “why” belongs to a strategy deck that was written for a fundraise and hasn’t met the sprint board since. When the why has no owner, the backlog fills by gravity.
The test that exposes it
Ask five people in your company — a founder, a PM, two engineers, someone in sales — the same question: “Why is the top item on the roadmap the top item?”
You’re not checking whether the answers are right. You’re checking whether they’re the same. Five matching answers means direction exists and is understood; you likely have an execution business, and execution is buyable. Five different answers means every day of delivery is being spent against a decision nobody actually made — and faster delivery makes it worse, in the precise sense that you’ll produce more unneeded software per unit time.
What fixing it looks like
Not a reorganisation. Three habits, installable in a quarter:
- Every roadmap item names its number. Not “improves experience” — the metric it moves and roughly by how much. Items that can’t name one go to a separate list labelled honestly: bets, compliance, or favours.
- Someone owns “no.” A written list of things you’ve decided not to do, with reasons. If nobody can name what you’re not doing, you don’t have a strategy — you have a queue.
- Review outcomes on the same cadence as delivery. The sprint review asks “did it ship?” Add the meeting that asks “did the last three things we shipped do anything?” — and let the answer reorder the backlog.
None of this is exotic. It mostly doesn’t happen because everyone senior enough to install it is busy, and the deadline is visible while the drift is not.
That gap — between shipping and mattering — is the entire reason Formula Station leads with direction work before selling you a team. But run the five-person test this week regardless of who you hire. The answers are usually educational within the hour.