How can I generate a timeline for tasks based on "typical man-days?"

Question

I need to come up with the planned and actual start & end dates for the following tasks below. How should I do it and what is my basis?

1. Finalize business user requirements
2. Technical Implementation Plan
3. Baseline Plan Released
4. Design
5. Development
6. SAT
7. UAT
8. Business Readiness
9. Cutover and Shallow Testing
10. Warranty and Closeout

Answer 1

Josh's comment is spot on, estimates are going to have the most value if they are provided by the people doing the work. There are some other things to consider:

• People can be unduly optimistic or pessimistic. You should look at actual durations for the activities on past similar projects for guidance on how long they typically take. Be aware that you may be comparing apples & oranges, but you need to do your due diligence to make sure the team isn't either saying what they think you want to hear or are padding their estimate unnecessarily.
• You should work off of multiple estimates. There are a number of ways to do this. For example you could get best/most likely/worst case estimates for each activity and derive a weighted average (e.g. PERT) to plug into your schedule. Or you could play planning poker. Or whatever else turns your fancy.
• Report start/end dates as ranges. Saying something will start/end on date X is largely a fantasy because estimates are just estimates. While you could reasonably fix a hard target for the final deliverable to your client, the start/end dates for the interim deliverables are going to have some amount of error. By providing date ranges you are in a better position to manage expectations.

Answer 2

Project Planning and Estimation

There's no boilerplate answer to your question, because planning and estimation are processes which vary widely between organizations and implementations. However, regardless of your methodology, there are some basics that you need to educate yourself about.

1. Identify resources.

   Figure out who is going to be performing the activities in your project plan, what equipment will be needed, and so forth.

2. Ask task performers for estimates.

   Depending on your methodology, this may take the form of time estimates, story point estimates, or cost estimates. Whatever metric you're using, you apply it to your work breakdown structure to build an estimated timeline.

3. Identify target dates.

   Using your estimated timeline and your management targets, you define target dates for specific things. This may take the form of a milestone, critical path node, or release date depending on your methodology.

Ultimately, the basis of your calculations will be your team's past performance on similar tasks, or educated guesses about the level of time or effort required for each planned task. Ideal man-hours are exactly that: an ideal. How many real or ideal man-hours required for any given task will vary widely, which is why detailed estimation of clearly-identified tasks is a necessary part of the process.

Estimation Exercises

Every methodology generally has its own set of tools for performing estimation. Scrum teams generally use some variant of Planning Poker to generate consensus estimates for defined units of work (generally called "user stories"), but there are certainly plenty of alternative methodologies.

Pick whichever estimation exercise gets you the most organizational buy-in. Just remember to remind people that estimates are not iron-clad guarantees. That will save you a lot of trouble down the road.