Recently I got a chance to work with scrum teams (who were new to scrum) and the management insisted that this project has a fixed end date, with fixed scope and a fixed budget. None of these can be changed. Because commitments have been already made to customer!

Following are the things that we tried and seem to have helped the project stakeholders.

1. Initially we had 3 week sprint length. We decided to go with 1 week sprint. Following being the reasons –

1. With 3 week sprint, managers used to pressurize teams to squeeze more work (you can imagine how this works with fixed price projects!)
2. In 1 week, very small work can be picked – thus controlling manager’s strong desire to squeeze more work
3. Requirements, people, technology everything was either changing or new!
4. In 1 month we could realize 4 times that we can move only with certain speed and deliver only so much software

2. Many impediments came out (some repeatedly) which helped to shift focus on idle work and not idle workers!
3. By failing faster (every week), we realized that we should be practical, because hope could be the worst strategy!
4. In about 1.5 month we had enough trend (empirical evidence) to acknowledge that we need to continue planning and throw out the plans(every week)
5. And this helped us acknowledge that our initial commitment (plans) of  how much  we can deliver in how much time using how many resources (cost) was how much wrong!
6. At least there are discussions about “what to do now?” J , which I think is still very much beneficial for the project!
7. Starting to think early about possible corrections, is far better than discovering bad news towards the end!

We may not be using perfect scrum yet (it is a journey and not a destination anyways). But, I think the iterative approach is helping with frequent feedback, opportunities to inspect and adapt. It’s now up to us, how we want to use that learning and respond to the situation at hand.

So, I think what Martin Fowler has said is very true. He has mentioned – “When should we use iterative development? You should use it only on projects that you want to succeed”

And if we want to succeed then we must learn from frequent feedbacks and adapt our plans based on practical experiences and not rely only on hope!

And I think iterative approach will help because what Ken Schwaber has said about scrum – “Scrum is like your mother in law, it will point out all your faults, every week”

Finally, we must understand that bad news early is a good news!

  –  चिंगुडे