The size of a work package always seems to create a lot of discussion. My typical response to queries of size becomes another question to be answered. The question I ask is, \u201cTo what degree of accuracy do you want to measure your project?\u201d If you want your cost to be within 5% of variance then you must plan each work-package to a level where you feel comfortable that you can estimate the cost to within 5%. The same holds true when considering the amount of time that each work-package will take to complete.There is no doubt that the smaller the work package becomes the more accurately we can estimate the work. This is the whole premise behind the creation of a work breakdown structure and the basis for the value added by sound project management practices.But there is another issue with which we must contend. That is our ability to measure each work package once the project gets moving. The question here becomes, \u201cHow long do I want to wait before I realize that I am late?\u201d Let\u2019s take the example of a work package that is planned to complete in 10 business days. When we estimated this work we were confident that we had estimated this task to within 10 percent. Therefore we expect this task to take no less than 9 days and no more than 11 days to complete.To be successful here we must be able to confirm the actual start of the work. We must also be able to measure the percent complete as we execute the work. If we are not confident in our measurement then our estimated variance can exceed the 10% we originally planned. So when we do our planning it is just as important to determine our measurement technique as it is to determine our estimates. Often our measurement technique within work-packages, even for a deliverables based plan, is merely by \u201casking\u201d the people doing the work, \u201cHow well are you doing.\u201d This then becomes an issue of trust and the savvy PM knows who to trust and who they need to watch closely. Not everyone can be trusted 100%. I might even go as far as to say that no one can be trusted 100%.\u00a0 There are a few that can be trusted much less than 100%. We all know them, we see them every day.So I go back to the original question, \u201cHow long do I want to wait before I realize that I am late?\u201d In the worst case scenario the work package should be no longer in duration than the amount of time which we can afford to be late. Therefore each work package would complete before we are beyond our limit of being late.There is much more to this topic but this is where I will leave this topic for today. I will create a follow up with more thoughts for next week. Send me your thoughts so I can incorporate those in my next posting.