Lessons from the Office of the CIO, Washington State
••11 min read
Written by Michael DeAngelo and originally posted in the Holacracy Blog of Washington State
The Office of the CIO has been operating with the Holacracy self-organizing model for part of the organization since February of 2015. On July 1st the same year, a law passed that consolidates our office with two other organizations in state government now known as Washington Technology Solutions (WaTech). This transition provides a great opportunity for a Holacracy retrospective over the past five months and a kickoff for the next phase of Holacracy in Washington government.
I mainly want to cover what the experience has been and lessons learned over the past five months for this post, but I’ll also touch on what what’s next.
Talking about Holacracy inside the organization
Over the course of the past five months I’ve spoken to countless people about self-organization and Holacracy. I’ve spoken to government leaders to create top down support by describing how self-organization may help position the state as a competitive employer in the region. I’ve spoken to peers to gain support and share my excitement. And, I’ve spoken to staff to test interest and build bottom up support. One of the things I’ve learned is that the reaction pretty much falls into one of two categories of reaction. The first is one of amazement. Wide eyes, piqued interest, and statements like: “That’s amazing!”, “Wow, I love it!”, or “That’s interesting” followed by a series of probing questions. The second type of reaction is best described as “What? I don’t get it.” About 80 percent of the reactions were in the first category.
The one thing I learned is that terminology can be a barrier and an immediate turn-off. Using the term “Holacracy” to describe the idea usually caused people to get stuck and prevented them from listening and learning about the concept. I switched my language to describe the idea simply as implementing a self-organizing and self-governing model without mentioning the word Holacracy. Self-organization and self-governing are familiar words that allowed me to continue the conversation to get to the value proposition. When I eventually introduce the term Holacracy it’s usually by saying something like “There are many self-organizational models out there–one of which is Holacracy which is package-able, repeatable system, currently implemented successfully in hundreds of organizations.”
My advice to anyone wanting to implement Holacracy is to take a similar approach, so that you’re not also fighting the language problem of unfamiliar terms before you can get productive conversations started.
The changes in authority created challenges
Within the test group, everyone struggled in one way or another with the changes in how authority is handled in Holacracy. For me, as the manager ceding my authority, I struggled with many of the governance proposals offered by staff because I thought they were suboptimal. As a manager my desire is to help staff build a better proposal, so I was always aware of other issues implementing an idea might encounter. However, in a Holacracy context, raising objections for this reason diffuses the ownership of the proposal and implies that the person is unable to process their own tensions and undermines their own learning. The new habit that I had to develop was to focus on processing the tensions of my specific roles, offer advice through the normal governance process, let go of my anxiety around sub-optimal proposals, and trust the people and the process. If the proposals are truly sub-optimal, the system will take of them by identifying the gaps quickly.
Holacracy also gives roles full authority (minus a few things) to energize the roles as you see fit. You are effectively the CEO of the roles you fill. People handle this new freedom in a variety of ways.
The most common challenge was seeing them coming to accept they really had the authority to take action and they did not need to rely on a manager for permission to get things done.
It’s hard to not the think of the Lead Link as the manager. First, the name implies it. Secondly, it’s usually the manager that initially fills the role. Thirdly, the role has accountability for budget, priority setting, and strategy. These are typically areas owned by managers.
There were two instances where the staff had tensions but wanted a traditional manager to make the call because it was the type of thing a manager would have done in the past. In both instances, anyone in the circle could have taken the action. However, some staff schemed to come up with a way to force the Lead Link to do the work.
The idea was to create a role in governance with this accountability knowing that no one would actually fill the role. In Holacracy, unfilled roles fall to the Lead Link. When this passed in governance the conspiring staff cheered in a tongue-in-cheek fashion. The strategic flaw to their plan was in not knowing that each person prioritizes across all their roles as they see fit. So, the Lead Link could simply prioritize this work to the bottom.
In a fully matured Holacracy, someone who recognizes needed work that is not specifically in one of their roles but serves the purpose of the organization, would have simply taken individual action then proposed a way to address this systemically through governance. However, the idea of taking individual action, especially outside your own role, is a muscle that is not often exercised. In fact, you can say that taking actions outside your role is explicitly discouraged in a hierarchy. We are still learning to use this muscle.
To help shift the thinking we raised tensions in governance to create roles that effectively move accountability and work away from the Lead Link role. For example, we now have the BizHub Budgeteer, which has authority over budget, and a role called Business Strategy, which has accountability over circle strategic planning, and our Culture Coach, which has accountability over the employee selection process.
These are still challenges that we’re working through. All I can say is that it will take us more than five months to unlearn some of the habits that we’ve learned through an entire career of institutional training.
My advice is to simply know that different people will take the authority shift differently. This will manifest itself the mostly in the tactical and especially the governance meetings, so having strong facilitators and coaches in the meetings will help identify and correct the bad habits and behaviors.
What staff have been saying
“Since I’ve been here certain problems weren’t getting solved and there was no forum to do so. Now there is.”
“It gave me a place to create transparency on an issue that I needed resolved.”
These sentiments have been repeated by several people. We’ve always had staff meetings, of course, and both I and the State CIO like to think we’re quite approachable individuals. As I dug into this more I began to understand what employees meant. The “certain problems” referred to were structural problems that were complex, impacted other positions, would have created expectations of other staff, and in the past, would have implied that managers weren’t doing their job. Staff meetings, at least ours, were not the right place to raise structural issues or at least not the right place to iterate on them. The staff meeting contained too many people and covered too many topics to deal with the issues causing tensions like these.
In both instances above, the individuals were talking about different issues which now have a space where these types of conversations can take place in a non-confrontational way. Holacracy supplies a place where everyone has equal authority, a place that is safe, a place with the specific purpose of raising these issues, and a place that can iterate on solutions.
“We are having project success because of Holacracy. There is now a system for addressing systemic issues.”
“Holacracy has improved our Scrum discipline.”
In the OCIO we apply agile principles and the Scrum methodology. One question that came up early on is whether Holacracy would add any value to a team that is already empowered through the Scrum process framework. What we learned is that even though a team is organized in Scrum teams around projects, they exist in a larger organizational context that can impact the Scrum team. The Holacracy structure allowed the team to manage the boundary between the team and the rest of the organization, as well as provide a venue for dealing with issues that really fall outside the project Scrum context– similar to the concept of the Scrum of Scrums.
In addition to the boundary management and issue resolution, the team also began adapting their Scrum methodology as a result of Holacracy. The team borrowed the tactical meeting format and added it to their toolbox. This formalized how bigger issues, raised during the standup, could be resolved quickly without dragging out the stand-up.
“Holacracy has ruined me. I no longer want to work in an organization that practices hierarchy.”
This sentiment was repeated by a number of different people. Whatever issues still existed or emerged, staff felt they had ownership of the issues. Additionally, they were conveying much greater confidence that systemic issues impacting their ability to perform would actually get resolved using Holacracy. Perhaps more pointedly, under a hierarchy the systemic issues impacting their ability to perform would likely have never been voiced, nor resolved.
For the past five months, the experience we’ve worked through was intended to allow learning about Holacracy and create a platform for a conversations about having a sustainable Holacracy model in government. However, we also collected data during the experience.
From the beginning we collected metrics based on the assessment model used by ScrumInc for tracking the performance of agile teams. We focused on two of the five metrics that were the easiest to measure in the timeframe of the experience: velocity and happiness. In this case, since the experience was about learning, velocity equated to the velocity of resolving issues (called tensions).
The happiness metric was used to measure the team’s confidence that Holacracy was helping resolve impediments to their work. It was measured on a scale of one to five using the agile “fist to five” polling technique asking the question: “How happy are you with your ability to get your role’s impediments resolved?”
The happiness metric started high corresponding to the general optimism about doing something new and faith in Holacracy’s ability to help. It immediately dropped. Staff become frustrated with the weight of the process of Holacracy. Over time, as the staff became more proficient with the rules, the process weight became less of an issue.
In general, there was a drop in the score each time a new member was added to the circle because they often voted lower with skepticism than the rest of the team.
The large drop on 5/29 was a reflection of the angst from the impending organizational merger that was only a month away. Many questions and tensions raised related to the consolidation that could not be processed in this circle.
Having said that, in general, we saw a steady increase in our assessment of Holacracy’s ability to resolve impediments to getting the work done. If the outlier on 5/29 is ignored, the trend goes from roughly 3.4 to 4.1 happiness level on a 1 to 5 scale.
There are two types of tensions in Holacracy: tactical tensions and governance tensions. Tactical tensions are very much akin to the sort of operational issues any organization deals with and commonly addresses in standard daily, weekly, or monthly operational meetings.
Our ability to process tactical tensions rose steadily throughout the experiment. An increase in the number of tactical tensions process shows an improved ability to use Holacracy as a tool to process operational issues and growing comfort with the Holacracy process. We know, however, this information is inconclusive as an indicator of Holacracy being a better system to resolve more operational issues than we did before since resolution of operational issues wasn’t measured prior to our experiment with the self-management model.
Governance tensions are a different beast, and the ability to resolve these kinds of tensions quickly, if at all, is often lacking for employees in many organizations. Governance is the management of roles, accountabilities, domains, and policies: in other words, the design of the organization itself. There isn’t an equivalent system in a hierarchy, since the design of the organization is largely reserved for those at the top of the authority pyramid. But in self-organizing systems, the employees decide how to organize.
Therefore, simply identifying and resolving one governance tension is infinitely more than what we were able to do before. For both the tactical and governance tensions, we saw a trend that increased over time to eventually three times more velocity than at start.
Anecdotally, this is what it felt like as well. We became more proficient at creating proposals that were implemented, but more importantly, we learned that proposals are incremental. Since there is an ongoing space for the process, participants learned that perfect proposals were not needed. The process occurs frequently, it’s really easy to fix mistakes, so there is less fear and overhead trying to craft the perfect proposal.
We also got better at identifying valid objections, which allowed us to moderate internal emotions about a particular proposal and become less emotionally vested in issues that did not directly impact our own work. As a result, our circle members became far less interested in being an impediment to organizational change and growth. This is an area where we still opportunity for improvement.
So what’s next for Holacracy in the newly formed WaTech agency? What we learned from the experience is that Holacracy can make a difference. We have yet to identify any reason why Holacracy wouldn’t work in government other than a lack of courage to try it.
In government, authority is shared by multiple branches of government rather than a single CEO. There are a lot of people who can torpedo ideas—even good ones. In addition, all our activities are open to public scrutiny and opinion, which can create a lot of caution within public organizations about being caught haplessly in “above the fold” situations, a term that refers to showing up in the front page of the newspaper because of an action reported as scandalous or wasteful. This can create a culture of perhaps overly conservative action with no upside to being bold–only risk for doing so.
Given that, for Holacracy to be considered as a viable and sustainable organizational model in Washington State government, there has to be hard data that demonstrates it adds value to the citizens in some way. Without a precedence in government, we’ll have to create that data.
The new agency, WaTech intends to continue the Holacracy journey by implementing a scientific approach to assessing the value of a self-organizational/self-governing model. We have partnered with the Harvard Business School to develop a study design that will test the hypothesis: “Holacracy produces better employee outcomes than hierarchy.” There are mounds of research showing strong employee outcomes improve customer service, organizational outcomes, and improve recruitment and retention measures. These are our goals here.
I’ll cover more about the experiment and study design in a future post.