Critical path: knowing which delay actually matters

On any plan with dependencies, some delays move the finish date and most do not. The critical path is the longest chain of dependent tasks through the work — and because it is the longest, it has no spare time in it. A day lost on that chain is a day lost on the project; a day lost anywhere else is usually free. Most status meetings do not know which is which, so they chase whatever went red.

Finding the critical path without software

• List the tasks with durations, and for each one what must finish before it can start.
• Walk forward through the chains, adding durations, to get the earliest finish of the whole thing.
• Walk backwards from that date to find the latest each task could start without moving it.
• The difference between earliest and latest start is that task's float. Tasks with zero float are the critical path.
• Expect the chain to be shorter than you feared — on most small projects it is five or six tasks out of forty.

What float actually means

Float is spare time, and it belongs to the chain rather than to the person doing the task. Two tasks sharing ten days of float do not have ten days each; the first one to use it leaves none for the second. This is why "we had a week of slack" is so often followed by a missed date — the slack was spent three times by three people who each believed it was theirs. Float is a project-level resource and should be tracked as one.

The critical path moves. Finish a critical task early or let a near-critical chain slip by two days and the path reroutes through different work. A path identified at kick-off and never recalculated is a historical document, which is how teams end up protecting tasks that stopped mattering weeks ago.

Using it to run the work

1. Put your best people and your attention on the critical chain rather than on whatever is loudest.
2. Ask for warning of slippage on critical tasks before the date, since those are the only ones where early warning changes anything.
3. Buy time by shortening critical tasks — more resource, reduced scope, parallel work — and ignore efficiency gains elsewhere, which produce nothing.
4. Recalculate after any significant change, because the path is a property of the current plan rather than of the project.
5. Protect float deliberately: decide who may spend it rather than letting it be consumed first-come.

Why most plans never find it

Because the dependencies were never written down. Durations are easy to estimate and everybody does it; the question of what genuinely cannot start until something else finishes requires a conversation that feels unnecessary until the week the date slips. A plan with durations and no dependencies cannot have a critical path at all — it is a list of tasks with dates attached, and its end date is a guess with arithmetic on top.

Ettex Board holds the dependency between cards, so the chain that sets the date is visible in the work itself rather than in a separate schedule. For anything long enough to need milestones, the same information decides which due date is worth defending.