The timeline should depend on external deadlines (such as funding applications) as well as the number of clients served (for programs that are already in operation). Before a program has started the timeline is largely a function of personnel and other resources. It is important to leave time in between iterations of tools used to measure outcomes, so the measures can be tested and refined. 

For programs already in operation, the number of clients is a key feature of the timeline — you need to be able to test the new/altered measures with enough clients to generate reliable information. The subsection on testing will give more details on this issue.

Different measures may also have overlapping but not synchronous timelines (may have some in the field while others are being tested). Using questions that have already been validated can drastically shorten the timeline, meaning it may take a few weeks to decide on measures, a few to test them in the setting (for at least 10 clients), and then additional time to change documentation.