A different way to group decisions
Do you know that situation? You had a group discussion, talked about a problem, and voted for a solution. Everybody seemed happy. But later, you find out that the vote is largely ignored, and the group doesn’t stick to the decision.
If that happens to you, you might be voting, but you’re not reaching a group decision.
In this post, I try to explain what’s happening, and offer a different mechanism for reaching group consensus.
For brevity, I’ll skip questions like ‘why vote at all?’ or ‘when is a vote better than a managerial decision?’.
What voting systems are there?
Do you know that old joke, ‘Five lawyers, six opinions’? The same can be said about voting and consensus-generating systems. Many of them are very elaborate and designed to steer large groups of humans through projects, economic decisions or political representation processes.
For this post, I’ll describe and tackle a few very simple ones that are very often found in workshop settings, meetings, or informal group gatherings. I then offer a new option to try out if you want to.
The Setting
To make the discussion less abstract, let’s consider a common setting: A project team wants to go to lunch together, and needs to find a place to go to.
The rules are very simple: The group has compiled a list of lunch places in the vicinity of the office. The need to decide on which place to go, and everybody has their equal say in the vote. There’s not much discussion (everybody’s hungry), they just vote.
We have five participants, Alice, Bob, Carla, Debby and Emil, and they found five suitable places: A burger place, a french restaurant, a pizza place, Sushi, or a vegan option. All of them fit the basic requirements of space, distance to the office, budget, and opening hours. Objectively, all are valid options, so it’s just a matter of taste …
The decision systems
What systems do we want to try? In my experience, these are the ones you will find most often:
- Show of Hands: Everybody has exactly one vote to cast. Could be anonymous, but usually you just raise your hand when your choice is called.
- Point-based Preference Voting: Everyone gets a couple of sticky points they attach to their choice. The typical number of points available is (number of options — 1)/2. If you’re allowed to give one option more than one point is situation-dependant.
- Full Ranking: Everyone ranks the full list of options into a ranked list, with the most preferred option on top, and the least preferred one at the bottom. Point-based preference voting is an abbreviation of a full ranking.
The Group Preferences
For reference, here is a peak behind the curtain: The full list of actual preferences our team members have, in form of a ranking.
The Results
So, shall the group vote? Let’s start with the easiest one, the show of hands:
Bob and Debbie have it — they want Sushi. And they’ll get it, because no other option has two supporters.
Next: Point based ranking. Everybody gets two points to spend, and they are not allowed to put both points on one option (to discourage tactical voting).
Now, that’s interesting. All participants voted strictly rational: They gave one point each to their first, and second preference. And suddenly, Pizza is number one, and Sushi — the clear winner of round 1 — is suddenly tied on second place with burgers.
Okay, let’s keep going, and get more complicated: We now ask our team members to rank the available lunch options from 1 to 5, with 1 being the most preferred option to 5 for the least preferred.
Unsurprisingly, the change is only nuanced. Pizza still wins, and now Sushi’s a bit ahead of burgers. But since we can’t go to more than one place, that difference doesn’t matter.
To sum it up, here are our outcomes, based on the voting system we implemented:
If we vote by hands (one person, one vote), Sushi wins.
If we vote by ranking, Pizza wins.
We’re done, right?
So, we’ve voted, time to get lunch, right?
Well, let’s just stick around for a few moments. When we look at the voting behaviour, one thing sticks out: One person — Carla — voted for the vegan option whenever possible. As it turns out, our ficticious team member is a vegan, and can only have lunch at places where she can find something that fits her diet.
What happens to her?
Let’s have a closer look at her preferences:
Since she’s vegan, she ranks the vegan option as her favorite, followed by the french restaurant , burger and then pizza. Sushi comes dead last. No wonder, who wants to eat kappa maki all the time?
In any voting, Carla is off quite badly. Ranked voting generates a winner she has as number 4 on her list. And the simple vote is even worse for her, when Sushi clearly wins. In fact, she’s the only person who ranked Sushi at the last place.
So, did we reach consensus?
Carla would disagree.
Now, let’s abstract from the example for a second. It is pretty clear that Carla will not be happy with the outcome. How energetic will she be when it comes to implementing the group decision? And since everybody thinks they have reached a consensus, how will group members and her boss interpret her (in)actions?
Finding a place to have lunch might sound mundane (and I do know people who’d disagree), but what about all the other things your organisation decides on all the time? Can you be sure that everybody actually agrees to the results and stands behind the decisions? Or do you find out time after time that for some reason, the group deviates from the decision, moving back to something completely else?
So, should we all have vegan then?
No, definitely not. But allow me to introduce a different way of trying to reach a group decision: Consensus by resistance voting.
Resistance Voting
I learned about resistance voting at a seminar, and immediately saw its potential. It’s part of a larger consensus mechanism called ‘Systemisches Konsensieren’, and I didn’t find any hint on how to translate it into english. So I made up my own wording for it. Do let me know if you have a better translation.
In contrast to other methods for reaching group consensus, Resistance Voting is pretty easy to understand and implement.
For each option, state your resistance to that option on a scale from 0 to 10, with 0 being the least, and 10 being the outmost resistance you feel towards it.
Now, if we have our team members perform this exercise, we arrive at the following result:
Note that I introduced a new dataset — the internal resistance is derived from the preferences table, but I filled in some detail.
As you can see, a few things are visible now: Carlas extreme resistance to Sushi (which more or less translates into a veto). The fact that Bob hates vegan food and stops just short of blocking it alltogether. Debbie just likes Sushi and is ‘meh’ about everything else. All of this detailed information that is highly relevant to a functional group got burried in all the other systems. Now we surfaced it.
We also know now that French and Vegan have very high values of ‘Group Resistance’, meaning that every attempt to implement something including one or the other will lead to a lot of opposition from the group (good luck with that ‘vegan cuisine’ idea for the office christmas party …).
And — here comes another surprise — we have a new consensus on where we should have lunch: It’s the burger place!
Apart from learning more about the groups preferences and limits, there is another benefit compared to the other voting systems: General group satisfaction is higher.
When you should best use resistance voting, and when not
Resistance voting by definition creates a compromise — something everybody can live with. Don’t expect people to be overly excited about the results.
If you need a lot of energy coming out of the vote, and want people to take the lead in implementation, resistance voting might not produce the right mindset. But if the decision you have to make is controversial, or people have a lot to lose, or you need the support of the whole group, this version of decision making might be for you because:
- Everybody can publicly state their limits
- Everybody can see the limits of everyone else
- Everybody will get heard
- The group will not come to a decision that is outside of the limits of anybody
- Mapping out resistances can surface deeper-lying conflicts
Try it out
I know, examples like this always feel constructed to work towards a ‘surprising’ conclusion that actually is the result of careful number massaging up front. That’s of course because that is exactly what such examples are.
So, I encourage you to try it out the next time you want to come to a decision in a group: Instead of voting with your hands, or ranking with points, try to map out your varying levels of resistance and see what happens.
