Guide

Finish to start vs start to finish: everyday examples table

Finish to start (FS) means task B cannot begin until task A has finished: the paint must dry before the second coat goes on. Start to finish (SF) means task B cannot finish until task A has started: the night guard cannot leave until the day guard arrives. FS is the everyday default. SF is rare and easily confused with FS.

What is the difference between finish to start and start to finish?

Finish to start controls when the second task begins. Start to finish controls when the second task ends. That is the whole difference, and both names read in the same order: the first word is about the predecessor, the second about the successor.

In an FS link, the predecessor's finish unlocks the successor's start. You cannot serve dinner until it has been cooked.

In an SF link, the predecessor's start unlocks the successor's finish. The babysitter can only finish their shift once a parent has started being home. The babysitter works first, but the parent's arrival is the predecessor, because it controls the moment the babysitter's shift ends.

This reversal confuses people. The task that happens first in time is the successor in an SF link. If you name the predecessor wrongly, the link points the other way.

The table puts each link type next to a household or office example. Microsoft's guide to linking tasks in a project uses the same 4 names.

Link typeThe ruleEveryday example
Finish to start (FS)B starts only after A finishesLaundry dries only after the washing machine finishes
Start to start (SS)B starts only after A startsSubtitling starts once the live broadcast starts
Finish to finish (FF)B finishes only after A finishesClearing the table finishes only after dinner finishes
Start to finish (SF)B finishes only after A startsNight guard's shift ends only once the day guard starts

For a fuller definition of each type, see the guide to the 4 dependency types. The words predecessor and successor are explained in predecessor and successor tasks explained.

Why is start to finish so easy to mix up with finish to start?

Start to finish looks like finish to start backwards, so people often draw an SF link when they mean an FS link with the tasks swapped. "The new system starts, then the old system finishes" sounds like a sequence, so it gets drawn as an FS link from new to old. That is not quite right: the old system can keep running for as long as needed after the new one starts. All the link really says is that the old system must not stop before the new one is running.

A quick test helps. Ask which task's end date should move if the other task slips. If a late start for the new system should delay the shutdown of the old one, it is SF. If a late finish of one task should delay the start of the other, it is FS.

The GAO Schedule Assessment Guide treats straightforward finish to start logic as the norm in a well-built schedule, and expects other relationship types to have a clear reason.

How do you rewrite start to start, finish to finish and start to finish as finish to start?

Most of the rarer links can be expressed as FS by splitting a task or adding a short handover task. The schedule becomes slightly longer to read but much easier to check.

  1. Split the overlap for start to start. Instead of "subtitling starts when broadcasting starts", create "broadcast first segment" and let "subtitle first segment" start when it is done.
  2. Split the tail for finish to finish. Instead of "clearing finishes when dinner finishes", separate "clear the last course" and make it start when "serve the last course" is done.
  3. Add a handover for start to finish. Instead of "night shift ends when day shift starts", create a short "shift handover" task that waits for the day guard's arrival, and end the night shift with it.
  4. Check the dates afterwards. Read the new chain from start to end and confirm that no task now waits for something it did not need before.

The extra tasks also make the plan honest: the handover becomes visible work instead of a hidden rule inside a link.

What happens in a planner that only supports finish to start?

A finish to start only planner forces every rule into the rewrite above, which suits small teams that want links anyone can read.

tokidu, the project manager by Sinergia Barcelona, which publishes this article, works this way. Its only link is finish to start, called «Starts when … is done»: a task stays blocked until the tasks it waits for are finished, then unlocks itself with a notice. If a finished predecessor is reopened, the task is blocked again, and circular waits are rejected. There are no leads, so a task cannot start before its predecessor ends. Waiting a few extra days after a task finishes is planned for the Pro plan, which is not on sale yet. You can see how this fits with the other planning features on the home page.

Tools built for complex schedules do more. Microsoft Project supports all 4 link types, as its linking tasks guide shows, and Microsoft Planner adds dependencies and critical path in its premium timeline view. According to Microsoft's lead and lag announcement, checked on 29 September 2026 and cited here on 17 November 2026, editing lead and lag in Planner requires Plan 3. If your work genuinely needs overlapping tasks with leads, those tools do that better. The Planner dependencies comparison covers which features sit in which edition.

Frequently asked questions

Is start to finish the same as finish to start in reverse?

Start to finish is not simply finish to start in reverse. In finish to start, the predecessor's end unlocks the successor's start. In start to finish, the predecessor's start unlocks the successor's end, so the successor usually happens first in time, which is what makes the two easy to confuse.

What is a real example of a start to finish dependency?

A start to finish dependency appears in shift work: the night guard cannot end their shift until the day guard starts. Another example is a system switchover, where the old system must keep running until the new one has started, whenever that turns out to be.

Which task dependency type is used most often?

Finish to start is the dependency type most planners rely on, because it matches how people naturally think: one thing ends, the next begins. Scheduling guidance such as the GAO Schedule Assessment Guide treats it as the standard logic and expects other types to be justified.

Can every dependency be written as finish to start?

Most dependencies can be written as finish to start by splitting a task into smaller parts or adding a short handover task. The plan gains a few extra tasks, but each link becomes easy to read and check, and hidden overlaps turn into visible work.

Sources

  1. Link tasks in a projectsupport.microsoft.com
  2. GAO Schedule Assessment Guide: Best Practices for Project Schedulesgao.gov
  3. Advanced project planning with Microsoft Planner: Dependencies and critical path in Timeline viewtechcommunity.microsoft.com
  4. Introducing Lead & Lag in Task Dependenciestechcommunity.microsoft.com

Written with the help of AI from tokidu's real data and features, and checked automatically before publishing.

ShareXLinkedIn
← All news