Reading total float

What total float is

Total float is how many days a task can slip before it pushes the end of the project. A task with twelve days of float can run twelve days late without costing you anything. A task with zero float cannot slip at all: the day it moves, the programme end date moves with it.

Tasks with zero float form the longest chain of dependent work through the project. Protect them first.

Seeing it on the table

Float is a property. Add it from the + at the end of the header row, under Schedule & PMO, and every task shows its days of room. It behaves like any other column: sort it ascending to see what is closest to the edge, filter it to zero to list what cannot slip, or use the column filter's "is between" to watch tasks with one to five days of float, the ones about to become a problem.

It travels with the table into saved views, footer calculations and exports, and the export summary can total it by owner or by group.

How it is calculated

Depentra sorts the tasks into dependency order, so nothing is scheduled before the work it waits on has been resolved. Then it runs two passes over the network.

The forward pass computes the earliest each task could start and finish, given everything feeding it. The backward pass works from the end date backwards and computes the latest each task could start and finish without moving that end date. Float is the gap between the two.

Lag on a dependency is carried through both passes, so a link with a ten-day wait built into it produces different float from a plain one.

Why the numbers keep changing

Float is recalculated against the current schedule, including any delays applied, rather than against the plan you originally baselined. That is deliberate: the slack you had in January is not the slack you have in April, because every completed task, every slip, and every scope change moves it.

It also means float re-forms while you are running a delay simulation, so you can see which tasks lose their room under a scenario before you accept it.