Friday, October 11, 2013

Velocity Dip or Double Dip – What to do?

Joe is a new to agile software development.  A year ago, he attended training programs on agile methods and learned more through his sporadic reading on topics related to Extreme Programming, Scrum, etc. He did not get an opportunity to practice though - he continued to play the role of project manager in other non-agile projects.

As you know, 70% of learning happens through practice, hands-on experience or problem solving.  About two months ago he got that opportunity he was looking forward to - his first agile project started.  A month before the start of this project he completed a certification program which provided him additional awareness and may be some confidence on how to lead agile projects and work with agile teams.

His project follows practices from Scrum, XP and other methodologies. He is a rookie Scrum Master.  This is the first time he and his team members are putting several agile practices into practice. They are experimenting and exploring. They are trying to make sense out of their experience iteration after iteration.

Last week weeks ago, Joe met me.   He wanted to discuss about the teething issues in his project. He needed my suggestions or thoughts on such issues.

We met on time and spent the first five minutes in catching up with some general topics.  Then he started on one of the issues in his project.

“Velocity is pretty much the same as throughput.  Is that correct?”

“Yes. You can say that. In simple terms it is the units of work delivered per iteration. What has been your team’s velocity? How is it trending?”

“Not good.  We are not seeing any increasing trend. It is pretty much the way the global economy is. Initially it was a single dip. Now we see a double dip!”

He smiled and I did too as I continued with my questions.

“So, tell me about your team. How is your team moving forward? Are they …”

“Well, the product owner is anxious.  Well, he is upset.  Yesterday we had our iteration planning session.”

“Concerned about the double dip?”

“Yes. Every dip is a matter of concern! How do we handle this situation? What do you suggest?”

“How was the team retrospective? What has the team reflected on?  Do you believe you can sustain?”

“I think so.  Our product owner knows the root cause.  User stories are not stable.  We have to do a lot of back and forth in refining most of them. That is tough.  This impacts our productive hours during the iteration. In spite of this he expects us to maintain an upward trend in velocity. He asks pointed questions on our velocity trend. Tell me. Is it this double-dip bad? What should we do?”

This is when I took a pause and then assured him firmly that it is ok to have a double-dip.  Why not?   A dip in the chart is okey as long as you collaborate enough, be transparent, identify root causes, implement corrective actions, resolve issues and move forward in the right direction.   ‘Fail-fast-fail-early’ is something we need to encourage and feel happy about.

Next, I asked him, “What are the success parameters in your project? How do you measure success? Is Velocity the only parameter or do you collectively look at other metrics?”

He understood where I was getting into. With a nod, he responded.

“I understand what you are talking about. I think we have not established a common understanding on our success parameters. We do not have a sound governance mechanism in place. I have a fair understanding on what to do now. I am going to take this up and discuss with our product owner.”

With a smile, I ascertained his approach. I think he wants to solve problems,  improve and reach the destination.

Before we parted, I shared with him my thoughts on the signs of collaborative spirit and my experience on how to build rapport with a remote product owner.

Joe is confident. He is going to meet me again.  I will come back to you on that meeting soon.  He met me after couple of weeks to discuss about an interesting thing  -  read the next post to know what happened in our meeting

Related Posts:

No comments: