Slack and float explained

Look at almost any real project schedule and you will find that most tasks are not actually due when their bar says they start. They have a cushion — some room to begin late, or run long, before anything bad happens. That cushion is called float, or slack (the two words mean exactly the same thing; "float" is more common in formal scheduling, "slack" in everyday use). It is the single most underused number in a plan, and learning to read it changes a schedule from a list of deadlines into a map of which deadlines are real.

Float is the answer to a very practical question: if this task slips, who cares? For some tasks the honest answer is "nobody, for days". For others it is "the whole project, immediately". Float tells them apart, and that is what lets you stop treating every task as equally urgent — because they are not.

Where float comes from

Float falls straight out of the two-pass calculation behind the critical path method. A forward pass through the network gives every task its early start and early finish — the soonest it could possibly happen. A backward pass gives its late start and late finish — the latest it can happen without pushing the project's end date. Float is simply the gap:

Total float = late start − early start = late finish − early finish.

If a task can start on day 3 at the earliest but day 8 at the latest, it has five days of float. The arithmetic is that plain. What makes it powerful is what the gap means: that is five days the task can absorb — through starting late, running long, or some mix — before it stops being its own problem and becomes everyone's.

Total float versus free float

There are two flavours of float, and confusing them is behind a lot of scheduling surprises. The difference is about who pays when you spend the cushion.

Total float — slack before the project slips

Total float is how long a task can slip without delaying the project's finish date. It is the headline number, the one most tools show by default, and the one the formula above gives you. The catch: total float is often shared along a chain of tasks. If three tasks in a row share six days of total float between them, that is six days for the chain — not six days each. Spend it on the first task and the two after it inherit a tighter, recalculated float. Total float tells you the chain has room; it does not promise that room is yours alone to spend.

Free float — slack before the next task slips

Free float is the stricter, more local measure: how long a task can slip without delaying its own immediate successor — the very next task that depends on it. It is computed as the early start of that successor minus the current task's early finish. Free float is always less than or equal to total float, and it is the safer number to act on, because spending it disturbs nobody downstream. A task with three days of free float can slip three days and the next task still starts exactly when it always would. That is a delay you can take without a single conversation.

A worked example

Take three tasks, A then B then C in sequence, that together must finish by day 12. Suppose the forward pass puts A at early start 0 / early finish 2, B at early start 2 / early finish 5, and C at early start 5 / early finish 8 — so the chain naturally finishes on day 8, four days before its day-12 deadline. The backward pass spreads those four spare days as total float across all three tasks: each has four days of total float relative to the deadline.

But the free float is different. B cannot slip at all without pushing C's earliest start, because C starts the instant B finishes — so B has zero free float even though it has four days of total float. The four days only truly belong to C, the last task before the deadline, which can slip the full four days without troubling anything after it. This is the trap in one picture: total float made all three tasks look relaxed; free float reveals that only C can actually move freely. Slip B on the strength of its total float and you have just eaten into C's cushion without realising it.

Zero float means critical

When a task has zero total float, it has no room at all: its earliest and latest dates are identical, so any slip pushes the project's finish. Those zero-float tasks, chained together from start to end, are the critical path. Float and criticality are the same fact stated two ways — critical means no float, float means not critical. This is why the critical-path tasks deserve your scarce attention and the float-rich ones can wait; the whole logic is laid out in the critical path method.

Negative float — already behind

Float can go below zero, and when it does it is a klaxon. Negative float appears when an imposed deadline sits earlier than the schedule can naturally deliver. If a chain of work needs until day 14 but someone has committed to day 11, the tasks on that chain carry minus-three days of float — meaning they would need to start three days before they realistically can, which is impossible. Negative float does not mean "tight"; it means the plan as drawn cannot meet the date, today, before anything has even slipped. The only honest responses are to compress the work (crash or fast-track the path), cut scope, or move the deadline — pretending the negative number away just defers the reckoning. Once you are in negative float, you have crossed from planning into handling slippage.

How to actually use float

Float is not trivia — it is a set of levers:

• Prioritise honestly. When two tasks compete for the same person on the same day, the one with less float wins. A zero-float task always outranks a five-day-float one, regardless of which feels more urgent or whose stakeholder shouts louder.
• Level resources without moving the end date. Float is the room you have to shuffle work around to smooth out an over-committed week. A task with float can be slid later to relieve a bottleneck, and as long as you stay inside its float, the finish date does not budge. This is the engine behind resource levelling.
• Know which delays are safe. "The vendor needs two more days" is a non-event for a task with a week of free float and a crisis for one with none. Float lets you answer instantly, instead of treating every wobble as an emergency or, worse, waving them all through.
• Watch float erode. The most useful early-warning sign in any project is float quietly shrinking. A task that had eight days of float last week and two this week is heading for the critical path. Track the trend, not just the snapshot — falling float is trouble announcing itself before any date is actually missed.

A note on independent float

For completeness there is a third, stricter measure you will occasionally meet: independent float. It is the slack a task has assuming its predecessors all finished as late as possible and its successors must start as early as possible — the cushion that is yours no matter how the neighbours behave. It is always the smallest of the three and is often zero, which is why most tools do not surface it by default. It is worth knowing the term exists, but for day-to-day work, total float for the project view and free float for the task view will carry you. Float is one of those concepts that sounds academic until the first time it saves you from a panic over a delay that turns out not to matter at all.

Frequently asked questions

Are slack and float the same thing?

Yes, exactly the same. They are interchangeable terms for the spare time a task has before it delays something. "Float" is the more common word in formal critical-path scheduling and most software; "slack" turns up more in everyday and North American usage. There is no difference in meaning.

What is the difference between total float and free float?

Total float is how long a task can slip without delaying the whole project, and it is often shared across a chain of tasks. Free float is how long it can slip without delaying its own immediate successor, and it belongs to that one task alone. Free float is always less than or equal to total float, and it is the safer number to act on.

How is float calculated?

Total float is the difference between a task's late and early dates — late start minus early start, which equals late finish minus early finish. Those dates come from the forward and backward passes of the critical path method. Free float is the early start of the next dependent task minus the current task's early finish.

What does negative float mean and how do I fix it?

Negative float means an imposed deadline is earlier than the schedule can deliver — the work needs more time than the date allows, before anything has even slipped. The genuine fixes are to compress the critical path by crashing or fast-tracking it, reduce scope, or move the deadline. The number will not improve until one of those changes the plan.