A five-day simulation

You are Amber.

You're shipping the unified customer export. It's a good project. Everyone agrees it's worth doing. Six people have to act for it to work, your nearest shared manager with half of them is three levels up, and you have five other things on your plate. It's Monday morning. You have until Friday.

odds it ships—

Friday, 5:47pm

—

Every other week you could have had

There were 81 distinct ways to play those four decisions. I ran all of them. Here is where every possible version of you landed, and where you landed.

the other 80 weeks the week you played the best week available

The thing worth noticing

Four decisions. Every one of them defensible. The spread between playing perfectly and playing badly was — percentage points — and the distance between all of it and a project that simply works was far larger. You were never the variable.

What would actually have moved it

Cut the project so 3 people are needed, not 6—

Run it entirely inside one team (shared manager)—

Have it be your only real priority—

All three at once—

None of those are decisions Amber gets to make on a Monday. They're decisions someone made months earlier, when they scoped the project and drew the org chart. That's the headwind: by the time it reaches you, it is already weather.