What are the most effective methods/strategies which will help in the setting of realistic deadlines within projects?
2
-
1I am not sure there is such a thing as a realistic deadline. However, I am sure there are plenty of unrealistic deadlines. It would be interesting to see what others have to recommend & share about how not to set completely "unrealistic" deadlines.– Corporate GeekCommented Dec 11, 2011 at 15:52
-
@CorporateGeek, that perhaps would of been a better question for me to ask.– Darren BurgessCommented Dec 11, 2011 at 15:56
Add a comment
|
3 Answers
First you need to define what you mean by realistic. If you were to do the same project 1,000 times, you would end up with quite a range of finishes. For example, a project could end with finishes that range from 40 days to maybe as high as 100 days. That entire range is realistic since it was observed. Or, realistic might mean something that has a higher probability than 1/1000. But my point is, how you and your stakeholders define realistic is very important and should be your first strategic action.
Second, use estimating techniques that allow for you to arrive at that range with the probabilistic density that sits on top of it. Avoiding allowing estimators to use single point estimating. A team of estimators need to look at a work package and understand what might result if everything ran perfectly and what might result if trouble was around every corner and what will most likely result. Use historical stats if you have it. And run a simulation so that you can see the effects of on the total project if you ran it 1,000 times. Then, using your definition developed in step one, you can pinpoint the interval that represents realistic to you and likely represents the highest probability.
Third, understanding that distribution will allow you to educate your stakeholder and set appropriate expectations, that despite choosing the highest probable finish, there is still natural risk of exceeding that time frame.
answered Dec 11, 2011 at 12:14
-
Thanks @David Estimation methods was what i really needed to hear, thank you. Commented Dec 11, 2011 at 12:32
-
Doesn't this address scheduling, not deadlines? A deadline isn't when the project actually finished, but the date at which a project must be finished to achieve some measurable objective (value added, regulatory compliance, etc.).– Thomas Owens ♦Commented Dec 12, 2011 at 14:56
-
The OP asked how to establish a deadline, which implies that he has the opportunity to influence it, derived from some logical way of establishing it. This is the way I answered it. Sure, in many cases, the deadline is given to us, based on other factors that have nothing to do with realism of achieving it, but that is not what the OP implied. Commented Dec 12, 2011 at 15:06
If the business's preference is to have fixed time for a project, then I would use a method that allows for prioritized and flexible scope, so that when we reach the set deadline there is better chance we have 80% of the features 100% done, instead of 100% of the features, 80% done. I am assuming also that money/resources is also fixed.
Agile methodologies are great for this (ie. Scrum, Kanban)
These methods would also allow you to better predict what set of features would be available at the deadline if the team is allowed to find their rhythm.
answered Dec 10, 2011 at 20:51
-
80% of the features is a software-centric answer. Would not work on other types of projects, say an airplane. Commented Dec 11, 2011 at 12:14
-
Great! Then if you fix your scope you use your velocity to measure your true completion date. Fixed Scope. Variable Date. Commented Jan 7, 2012 at 0:02
The biggest suggestion I would make is to ask the people doing the work to make the estimate. When someone makes an estimate then they have a vested interest in living up to what they stated.
On the other hand when the estimate is made by someone far, far away with no input from the team the results are often disastrous. Think back to a time when a boss cam to you at 3:00 PM and said "Marketing just told a customer we would have this change done by 9:00 AM tomorrow" and it was a big, multi-day effort. Did you feel like working all night on it while the marketing guy went out that evening to a party?
