Title: A Guide to Group Decision Making

Author: Bob Schmidt

Date: 06 September 2018 18:36:13 +01:00 or Thu, 06 September 2018 18:36:13 +01:00

Summary: Glen Stark advocates an approach to promoting team harmony.

Body: 

When groups of people gather together to undertake a project, decisions have to be made that affect the group. There are a number of ways that these decisions can be made, some of them are better than others.

In my experience as a software engineer, I have found that the decision-making process is usually poor. The agile manifesto and agile processes have had a transformative effect on how software engineering is conducted. The Agile Manifesto is nine lines long. Two of those lines concern group decision making: “Individuals and interactions over processes and tools” and “Customer collaboration over contract negotiation” [1]. One of the twelve principles behind the Agile Manifesto is “The best architectures, requirement, and designs emerge from self-organizing teams.” [2] Clearly the creators of the Agile Manifesto were concerned with collaboration, but most ‘Agile’ processes scarcely touch, or completely avoid, the topic of group decision-making and group dynamics.

Given the inefficiencies, stress, and sometimes dire consequences that arise out of our poor group decision-making, one would think we would study the topic in our basic education. Sadly, I’ve never met a person who has had such a course.

This is my attempt to draw some deliberate attention to how teams are making decisions. I identify some good and bad strategies for decision-making and touch on the pros and cons of each of them. My criteria for judging processes are the time and effort costs of the process, the impact on the team’s health, happiness, and productivity (which are strongly correlated), and the quality of the decision achieved. I hope that readers will find obvious ways to integrate these thoughts into their existing processes.

Useful decision making strategies

These are the decision making strategies that I consider useful and productive for groups of people. All of these methods work well on a small scale – perhaps for a group of two to a couple of hundred people.

• consensus: The group debates a decision until concurrences is reached – everyone agrees the decision is the best one on which consensus can be reached. The chance of reaching consensus decreases dramatically as groups grow, but the process can still be useful.
• democracy: The group votes on the decision. There are a wide range of possibilities for democratic decisions, but the most common is simple majority.
• expertise-based authority: One of the group (or perhaps a subgroup) has expertise in a particular topic, so the group agrees to defer to the obvious expert.
• stalemate strategies: Consensus can fail to be achieved. Democracy can be foiled by even numbers. You need strategies for these cases, like coin tossing, taking turns, or resorting to an outside authority.

While each of these strategies deserves deep consideration, it is helpful to attempt an overview of how these strategies compare and interact. Perhaps the most important thing to consider is that each strategy has value and deserves to be practised.

In the event that an individual or sub-group has particular expertise on a topic, it often makes sense to delegate decisions on that topic to the experts. They will often be able to determine a clearly better outcome with little discussion. As soon as decisions have to made about trade-offs that affect other parties, however, those parties should be involved in the decision. When authority is overstepped, it breeds both resentment and poor decisions, so groups should use this strategy with care.

A team typically has a large number of decisions to make where the entire team has a certain amount of expertise. As far as I can tell, these decisions should be made using consensus when possible, democracy when necessary, and other approaches only in corner cases. Striving for consensus and democracy should occur frequently enough that they are familiar and comfortable tools. This usually means applying them to less critical decisions on a regular basis, so that the process is smooth when a critical discussion must be had.

I’m convinced that decision by consensus, when reachable, produces the best outcomes by far. When a group reaches consensus, everyone can be happy about the outcome or at least accepts the necessity of that outcome and feels like their voice mattered. Unfortunately, there is not guarantee that consensus will be reachable for a particular issue – it may be impossible or too costly to reach. When consensus fails, a good fallback is to rely on a democratic choice. There should be a clear plan how long to work on consensus, when to fallback to another process, and what process to fallback to. Usually the best fallback option is democracy.

When making a democratic choice, the usual – and usually best – approach is a poll for simple majority. When resentment might play a role, consider anonymous voting. There are many variants on the democratic decision-making process and it is worthwhile to survey some of these variants to see if improvements can be found for your process, particularly at larger scales.

Democracy is riskier than consensus, and riskier than decision by authority (when the authority has a legitimate claim to expertise). The risks of democracy include factionalization and resentment, which should be monitored and combated. In my experience, this tends to be more of a problem at large scale, and less of an issue within a small team, but if your experiences have been different please share them with me.

Putting this together, an obvious algorithm is reached:

1. Is there a clear and acknowledged expert on the topic? Then let the expert decide. Remember that experts can and should be challenged, and when the decision being made affects other parties they should be consulted and ideally part of the decision-making process.
2. In the absence of authority, go for consensus. Often decisions are pretty obvious and saying something like “It seems to me like X is a good choice, does someone object?” will steam right though since most people want to get out of the meeting room. When objections occur, let everyone have their say. You need a clear strategy to follow regarding how long you will strive for consensus, and your fallback will be when consensus fails.
3. Democracy should be the default fallback. Some people will be unhappy but, when done right, people will feel that the decision was fair, keeping morale high. In the absence of consensus or legitimate authority, this will produce the best outcome and is not costly.
4. There are rare cases where a person in a position of administrative authority, like a manager or team leader, might find it best to exercise that authority. This can be useful in extreme cases where the leader feels it is important for team morale, or has knowledge they can’t share that enables them to make a more informed choice. This should only occur rarely.

Harmful decision making strategies

Doubtless there exist countless identifiable patterns of harmful group decision-making. Some like authoritarianism, monarchy, cult of personality, and fear-mongering are well understood but nonetheless continue to play a large role in both the private and public sector. We can only try to be aware of these patterns and seek to avoid them when we have the opportunity. Some other common anti-patterns I’ve recognized, which are perhaps less discussed, are:

• administrative authority: An individual is granted an administrative function which gives them the authority to override their teammates and make decisions by fiat. This can be very time efficient, but has many vectors by which it produces unsatisfactory outcomes. It should be an unwelcome last resort, but is sadly often the default.
• avoidance of conflict: Often organizations seek to find ways to avoid having discussions and conflicts among their team members. When a decision is made, it is made in the background without consulting others for fear of difficult confrontations. This typically leads to the most difficult and stubborn team members leading by default.

One common anti-pattern I’ve noticed in scrum is having a team leader (an administrative authority position) be the scrum-master. This organizational shortcut often leads to the default decision-making process being decision by administrative authority. When it’s done well, the administrator is responsive to the input and opinions of the team, but I would argue it is better to separate the organizational concerns. My current thinking is that the administrative authority and the scrum-master should be different individuals, and the administrative authority should be consulted as the authority when it is determined that the decision to be made requires administrative expertise. When group decisions have to be made, the scrum-master can decide how long to spend trying to achieve consensus, and whether to default to authority or democracy. It should be clear to the team what the process and fallback are in advance. I think rotating the scrum-master position is also valuable.

Last words

In this article, I have formalized things which seem self-evident to me, but in my experience the logical steps indicated by these observations are rarely practised. My suggestion is for we citizens and employees to study these approaches, and seek to correct and improve whatever decision-making processes we can. Ideally, effective group decision-making will become more of an active topic of conversation as this attention can only improve our processes. I welcome any contribution a reader might wish to make to the conversation, especially if your view contrasts with mine.

I encourage all readers to seek opportunities to practise consensus-building and democracy whenever possible. Do this often enough and you may inspire others to do the same. At the very least, your colleagues will appreciate your team spirit and open-mindedness. As consensus-building and democratic choice are practised more in your organization, they should become more accepted. The positive outcomes will help to spread the methods. Pitfalls that have to be navigated include problematic colleagues, and excessive time-suck.

In my experience, following this advice leads to happier and more productive teams, and better decisions. As with anything worthwhile, there are challenges and intricacies to consider, but I hope you will agree that good group decision-making deserves attention and practice.

References

[1] Manifesto for Agile Software Development: http://agilemanifesto.org/

[2] Principles behind the Agile Manifesto:http://agilemanifesto.org/principles.html

Glen Stark Glen is a human being, working as a software engineer on planet Earth. He cares about people, simplicity, reliability and efficiency. He likes to solve problems, ideally without introducing too many new ones in the process. He can be reached at mail@glenstark.net.

Notes: 

More fields may be available via dynamicdata ..