Exit criteria refers to ‘how can you tell that an element of the plan is complete?’
This is so that you can ‘exit’ that phase and move on to the next.
Simple examples might be:
Have all necessary reports been completed, circulated and filed?
Has the boiler been tested and the release certificate issued?
Here completion does not necessarily mean it has been successful just that it has reached a particular stage that allows a report to be drawn up.
This will usually be the subject of quality control, for example:
e.g. 95% of units light (this is OK for a specification requiring > 92%)
e.g. 70% of calculators worked (failed specification as > 95% required)
At any milestone, based upon the information that is being reviewed decisions have to be made.
In terms of major project stages these decisions will be made by senior management.
This would be the Project Board or programme management under PRINCE2.
Decision points usually require one of three outcomes in simple terms these would be:
For each of these there will be variations.
For example, the decision of ‘Yes’, to go ahead, may be subject to modifications of the plan.
In any case, the decision to proceed will not be based solely on data from the current stage.
In addition, the Project Manager must put forward a detailed plan for the next stage for approval.
When the decision is ‘No’ it could lead to project closure.
Similarly, ‘Yes’ could lead to project closure if it is the project end point.
In both cases, it is not simply a matter of stopping the project many loose ends have to be tied up.
This is covered in more detail within the process 'Closing a Project (CP)' under PRINCE2 [see ‘The Complete Project Management plus PRINCE2’].
The decision point can be very difficult.
It is easy to view the results in a favourable light and convince your self that the odd negative result will somehow right itself in the next phase.
This is often the case when a lot of money has already been spent.
Hence, make sure that the exit criteria for milestones, on which progress decisions will be made, should be defined with care.
In general, exit criteria cover two distinct areas.
Those for individual tasks and those for the major objectives at the milestone.
If a 4 week piece of work has been going for 2 weeks it doesn’t follow that there are 2 weeks to go.
You have to be very wary when reporting progress. It could be that delays have been incurred or a problem exists.
If the project is run by exception and all tasks are on track for completion as predicted then the project will succeed on time.
This is assuming that the plan and schedule are complete and accurate.
But how will you know that a task has been completed?
Have reports been completed and filed? Has the end product been officially released?
Has the piece of work been a success?
Did a specification exist on which to make a judgment?
It may be impractical to specify ‘exit criteria’ for all tasks but they should be considered by individuals and recorded where practical.
In exactly the same manner, milestones need clear criteria to show that a particular activity is complete.
There should be no ambiguity in the completion of a milestone.