Blog

Podcast: Flying at Portfolio Level

First, there was personal productivity. Then, everyone started talking about Agile teams. Today, it’s time to talk about Business Agility.

In a recent podcast I talked with Dima Moroz of  Kanbanize about how one does get an organization to do the right things at the right time. The trick is to move the most effective levers and if you focus on the whole organization, you better set that lever as high as possible. Only in rare cases does it depend on only one (agile) team whether real value is delivered to the customer. Most of the time this is a matter of agile interactions between multiple teams. However, Business Agility only thrives when someone on the portfolio-level has done some thinking about how just so much work enters the organization that projects are finished on a regular basis.

Dealing with constant disruption

I am regularly asked a question during customer visits or training sessions: How do you deal with constant disruption? I have already written about this, such as the blog article on white noise and we cover the topic also in the book Kanban Change Leadership. For many years, though, I more often recommend the “Dude of the Week” approach.

The initial situation. A team constantly receives requests to “quickly take care of” a small issue. Naturally, this causes a few problems:

• You must interrupt work that has already been started, leading to an increased cycle time for this work.
• Interruptions also cause adherence to delivery dates of the interrupted work to suffer.
• The context-switch isn’t free either. Once you are mentally taken out of the work, it needs a certain amount of time to get back into it.
• With every small piece of work that you “quickly” take care of, you start an expedite process since the short work takes over all other work.
• Thus, a reprioritization is implemented.
• In most cases, eventual WIP limits will also be exceeded.

All in all, not ideal if you really want to work flow-based. However, it also cannot be argued that someone must wait two weeks for a quick firewall adjustment, for example, just because you are in the middle of something big.

A solution. For such cases, what I have established in many organizations is the “Dude of the Week”. This is one or more persons (according to team size and degree of specialization) who only work on the small tasks for a period of time (a week, for example). So, 100% of the small requests are taken over by these people and worked on immediately. The other team members are able to dedicate themselves 100% without disruption to their large tasks and will not be constantly torn away from their work. If a small request actually turns out to be a large one, a ticket will be created and worked on through the normal channels. This is crucial, otherwise everyone would just request supposedly small tasks.

The improvement process. I also recommend that the “Dudes” collect the requests on Post-it notes. These Post-it notes will be analyzed in an improvement meeting. Perhaps you find that certain requests are made over and over and you can come up with a systemic solution, such as automating the task or changing the work process so that the request doesn’t come up as much in the future. That’s a similar approach to Blocker Clustering.