Note
The plan that died in month two
Almost every contractor has one. A schedule that took a week to build, looked excellent in the kickoff meeting, got printed and stuck on the wall, and then quietly stopped matching reality somewhere in month two.
The usual explanation is that people did not follow it. That explanation is comfortable, and it is wrong.
Detail is what kills a plan
A plan has to be updated more often than the work moves, or it stops being a plan and becomes a picture of the past. So the real constraint is not how accurate you can make the schedule on day one. It is how much of it you can refresh in the time you actually have each week.
Two rules keep a plan alive:
- No task may be longer than the update cycle. If you review weekly, a three-week bar tells you nothing for two of those weeks.
- One bar is one person doing one kind of work. The moment a bar covers two trades, nobody can answer whether it is finished.
Follow those two rules on a normal project and you land somewhere around twelve to eighteen bars. That feels thin if you are used to a two-hundred-line schedule. It is also the difference between a plan that gets updated and a plan that gets laminated.
The percentage field is the other killer
The second thing that kills a plan is the progress percentage. Someone types 60%, everyone nods, and the number cannot be checked by anyone. Percentages are why plans look green right up to the day they break.
There are only two honest answers on a construction task. Finished, or not finished. And when the work genuinely comes in pieces, count the pieces: seven zones out of twelve, not 58%.
That is worth saying out loud because clients ask for the percentage field constantly. It is the one thing worth refusing.
What to do instead this week
You do not need new software to test this. Take one live project and try it for a month:
- Write the project as twelve to eighteen bars. If you cannot get under eighteen, your bars cover too many people.
- Delete every percentage field you have.
- Ask two questions a week, in the chat group people already use. Which zone finished, and what shipped from the shop.
- Spend thirty minutes on a fixed day reading it with whoever decides things.
If nobody answers those two questions for two weeks running, that is not a discipline problem to solve with reminders. It is information: the thing you built asked for something people cannot give you, and the design has to get smaller.