Craig Larman, co-inventor of LeSS (Large-Scale Scrum), paid a visit to wemanity offices on 19th March for an agile consortium event where he presented his views on scaling agile and enterprises optimizations.
Status quo for managers and experts
Craig’s session, punctuated by bad jokes he is famous for, began with examples of what could happen in organizations based on component thinking.
Component based organization would typically have teams built around applications or technologies. Often the idea behind such enterprise architecture is to maximize the individual performance of each team making them experts in one particular technology, for example.
Such organizations, in Craig’s opinion, would ultimately generate more cost and delay, lack of visibility, queues, multitasking, low quality and a lot of defects.
According to him, the first law of any organization is the following:
“Enterprises are optimized to keep the status quo for managers and experts.”
Because of this, priorities get adapted to keep the same component organization in place which results in keeping people busy rather than generating value. This approach is not necessarily wrong. It depends on what the organization wants to optimize for. For example, if the purpose of the company is to increase the security / secrecy level (e.g. in the army), component thinking could be a good way to maximize ignorance.
Craig Larman also explained that, in order to keep status quo, some experts might choose to maintain an illusion of complexity. This creates a culture with less motivation to change, learn, improve and pay attention to quality. Newcomers educated in such environment will learn slowly, and will hesitate to modify existing code due to its apparent complexity.
The next generation leaders
An interesting part of the presentation was when system thinking and global optimization were addressed. Indeed, too often, we see organizations choosing to act locally instead of thinking globally. One of my favorite examples is sales bonuses rewarding new leads / contracts, no matter whether the company can absorb the new business or not. Another example is scrum teams not helping each other to optimize their own velocity. By doing so, they are probably increase time to market for the whole company.
When running an agile transformation, there is also a risk to focus mainly on team members as they are the first to be directly impacted by the change. Craig Larman emphasized the importance to include managers in the process. If not done, it will set agile adoption to fail. How can one make sure managers are on board? First of all, they need to be reassured about their job security, and their new role in the transformed company should be explained to them. For instance, at all levels, managers can be elected by their direct reports. The role of a manager would then become to serve, or support the teams instead of managing them.
Craig Larman also stressed how important it is for managers to recognize code quality. This is what I would call the “gemba genbutsu” principle applied to companies that build products by coding. Basically, it is one of Toyota principles that says that managers should walk the floor and talk to people working directly on the subject, instead of having meetings far away from reality.
Perspectives for managers and experts? Leadership!
Although I was disappointed that the event was some kind of teasing to Craig’s training sessions without speaking about LeSS, I must say that I enjoyed this event. It made me think of the managers and experts transformation into leaders. Indeed, I believe that this mindset change is definitively required, in an agile transformation, to make Craig’s law obsolete.
Will the next generation leaders think globally, be aware of the floor reality and help their employees?
Hopefully we will build this reality together at our next events!