Bike-shedding, Bystanders and Boredom:
I have a theory that in the context of self-organizing teams in a business environment, there are zones of diminishing responsibility and thus the effectiveness of self organization as the tasks get further from the perceived core purpose of the team or individual.
That is to say that a self-organizing software development team may have little difficulty organizing how to create a feature in their product, but may struggle to water the plants or empty the bins in their Team area.
We as an organization have embraced Spontaneous Order (self-organizing teams) and have adopted a variety of Holocracy, the result is a pretty large organization (Approximately 400+ people) with almost no management outside of the core leadership team of four. The results are fascinating to observe, in some cases wildly successful and in other cases not so much.
Teams are created for a defined purpose, most often this is a defined project from a client, the composition of the team is decided by one of the leadership team based on personal knowledge of the skills of the prospective team members. This is obviously a limitation of scale and requires an exceptional level of knowledge of the available people. Clearly it is not self-organization in the context of team creation but this is the extent of outside influence for most teams (unless they have problems). The teams are expected to self-organize to deliver on their defined purpose.
For the most part they do very well and are able to deliver high quality software, interact effectively with clients and have a reputation for being the best in the custom App development business. Although there are a few internal difficulties, such as a tendency for excessive bike-shedding* early in projects, and for a few individuals, especially the stronger personalities to shape the teams to enable them to do the work they like relying on the others to do the rest.
Similarly strong personalities can limit a notion of continual improvement, that is to say that strong personalities can instill pressure to stick with what worked previously rather than being open to improvement.
Those are problems that could be explored further but for today my interest is in observing how wider organizational responsibilities or team responsibilities that are perceived as distant from the defined purpose get lost, ignored or simply take far longer than they should to get done.
*Bike-shedding (or Parkinson’s law of triviality)
Way back in 1957 Parkinson observed that members of an organization give excess weight to trivial issues, spending a disproportionate amount of time discussing issues that everyone has an opinion on (in his example it was the the building of a bike shed), this discussion was at the expense of more important issues that were outside the domain of expertise of the entire group. In other words people contribute because they ‘can’ not because they ‘should’.
In my observations this is a MAJOR problem for self-organizing teams, the choice of electronic Kanban board is my favourite pet hate on this, the amount of time spent deciding on which tool to use or which columns to include, or future discussions about whether to add or remove a column is astonishing, these same teams will spend hours on these topics but will then declare a story writing activity unproductive if the quantity of stories written in an hour is below an arbitrary number, regardless of whether the discussion was effective in identifying details valuable to getting a feature right. Please note I am not suggesting that the decisions about tools or workflow are not important, just that it should be proportionate.
To remedy this self-organization may need a helping hand, a facilitator to keep the group focused, or to redirect in the face of bike-shedding.
Structure vs Empowerment
I often hear people talk about not wanting a framework imposed on them (e.g. Scrum) or “wanting to find their own way”. But in my observations many do find themselves with a workflow that would be described as Scrum to an outside observer. The concern I have with this is balancing the freedom to find your own path – allowing teams choose their path, balanced with the waste involved in teams repeatedly churning for weeks on end or longer until they find the path which is ultimately very similar to what would have been prescribed in the first place. Is it wrong to build on past experience by suggesting a proposed framework structure? Is it empowering to allow teams to repeatedly re-invent the wheel? I am oft told that it is empowering for them, but I wonder if they would feel the same way if they were paying the bills?
This applies to other parts of the organization too, it is all very well when entering a restaurant to be offered no menu and told to order anything you want. But when faced with that how many would feel imprisoned by too many unclear options, how many would rather have a menu and feel empowered to stretch it a little asking for tweaks or combinations, in other words having boundaries that can be pushed.
Observations of children playing show that if a play area has a fence they utilize the whole area, but when the play area has no fence the children cluster close to the center with each other. In our subconscious we are more daunted by the lack of boundaries, but when we have them and feel safe and empowered we will push at them.
In retail, studies show that customers faced with 6 choices are 15 times more likely to make a purchase than those with 24 choices, too many choices confuse and frustrate us and so we regress to safety which is to NOT make a decision. The same is true in our business life, we become overwhelmed with choice, constraining options is empowering not limiting.
The three layers of self-organization responsibilities
The teams will generally have activities and responsibilities fall into three camps, either trivial issues get blown up in to long and largely pointless debates that drag on, more complex issues get glossed over leading to problems later and finally there are certain responsibilities that get overlooked or dismissed as unimportant or out of scope of the team.
These are tasks that are clearly and directly related to your role on a team and directly related to achieving your team’s goals and objectives. In the case of a developer this may be writing code for specific user stories. Please bear in mind that a named role such as Developer or Quality Advocate or even Product Owner may imply certain responsibilities are theirs rather than the team as a whole (I am observing this not endorsing this)
These are the tasks that the team are responsible for but may not be perceived as core for the individual, again in the developer’s case this could include organising meetings, testing or writing stories or deploying or… well you get the idea.
These are tasks that need doing and are essential to the function of an organisation but are outside the explicit responsibility of the team such as: emptying bins, washing dishes, watering plants. Or less trivially ordering stationery, updating company web pages, ordering IT equipment and so on.
Exceptions: There seem to be two exceptions to these layers, the first is if a task in any of the layers is perceived by the individual as interesting or rewarding in some way, in which case the layers can be overcome, the other is when someone of authority (including peer pressure) explicitly assigns responsibility to an individual or small group which elevates the task to a Primary task.
Bystander syndrome is a phenomenon of anthropology where the more people are responsible for something the less responsible the individual feels. Sadly this syndrome is the reason that self-organization so often fails. Why in small teams or small companies things work well but as they grow this attitude slowly creeps in until one day there is a pile of dirty dishes in the sink, un-emptied bins, and someone has stolen my chair because their’s is broken and it was quicker to take mine than order a new one*. (*imaginary scenario – honest)
These minor annoyances are the start of the breakdown in your corporate society and the warning signs that self-organization has reached it’s limit. To keep things running new roles will need to be created to make people explicitly responsible for these Tertiary tasks. In short you will no longer be self-organizing someone will need to lend a discrete hand, to enable self-organization to continue where it really matters.
Essentially these are growing pains, the result is what can be best described as a failing of human nature. The tasks that get the most effort and attention are those that fall in the sweet spot of being close to the core purpose of the team, are trivial so that everyone has an opinion, and are interesting or appealing or have clear ownership. The tasks that get ignored and overlooked are those that whilst important or necessary in the grand scheme do not map directly to the core purpose of the individual, are uninteresting, or have no clear owner. For these self-organization wont save you, a helping hand (or a gentle prod) is needed.
Providing self-organizing teams and people with boundaries and structure and for our leadership to have an understanding and acceptance that we thrive best when we know our boundaries makes for a safe environment to grow. We will push at those boundaries mercilessly if we feel safe. But don’t be surprised if the plants don’t get watered unless you assign the job explicitly!
4 thoughts on “Why don’t developers water the plants?”
100% conforming to my experience in multiple organizations. Self-organization can not be expected for everything, at all scales and from every individual.
This is one of the best articles about bikeshedding I have ever read, thank you!