The accidentally selfish team
Picture your team, for a moment, sitting around a table.
On the wall is a timer, counting down from 2:00.
In the center of the table is a big red button.
The game runs like this. The timer will tick down until it hits zero. If no one has pressed the button by then, everyone gets a measly $1. But if someone does press the button, that person gets $1.25, a small bonus, and everyone else gets $2, a big bonus. After the timer hits zero, it resets back to 2:00 and the game starts again.
Now, picture your team again.
Who hit the button?
The button game is not something I made up. It was part of a research project, which you can read about in full here ( https://hbr.org/2018/07/why-women-volunteer-for-tasks-that-dont-lead-to-promotions ). It was a fascinating project, and a well written article. It also talks about some of the gendered patterns that they identified, and goes on to examine what happens when managers have the option to choose individuals in their team to volunteer. It’s definitely worth a read.
But ultimately, that research is focused on large, cultural trends. Let’s stop for a moment, and look specifically at your team.
There are a couple of ways this game can play out. The simplest case is that everyone stubbornly refuses to act. After ten games, everyone walks away with $10. It’s fair, but it’s a pretty poor outcome all around. In a team of five people, that’d be $50 total. As a manager, if my team was producing the absolute minimum and no more, I’d be pretty disappointed.
In the ideal case, each person would press their fair share of buttons. In that same team of five, everyone walks away with $18.50, nearly double what they would have had otherwise. Collectively that’s over $90, something a manager could be proud of.
Then there’s the third case, where one person becomes the default volunteer, and everyone else reaps the rewards. The collective team comes out ahead, but that one volunteer… Well, you can see where this is heading.
In an engineering team, the rewards of the game are not quite so clear cut, but the situations still play out frequently. The only difference is the shape of the value.
For example, consider a build which is intermittently failing because of a flaky or unreliable test. If it continues to fail, everyone slows down, but if one person fixes it and the build runs smoothly every time, everyone benefits. Less time wasted in rebuilds, less effort wasted investigating false alarms… But the person who stopped to fix it has made less headway on their assigned task than all those who just hit rebuild.
Who is the person in your team who stops their work to investigate and fix the underlying build issue, and who is the person who hits “rebuild” and leaves the real problem for someone else?
If a medium priority bug is found, who volunteers to investigate and fix it? Who helps the new developer set up their computer properly so they can start coding? Who upgrades the out of date dependencies before they become security vulnerabilities?
These people are your volunteers. They care more about making your whole team effective than they do maximizing their individual successes. They will take the $1.25, even knowing that everyone else will get more, because it’s better than everybody getting less.
They are fantastic people to have on your teams, for obvious reasons, and if you had an entire team full of them you would be flying. But when it comes time to assess them for recognition, rewards, promotions, andpay increases, you look at their KPIs. Their ‘dollar total’.
“Only $12.50? From ten whole rounds of ‘press the button’? You know, everyone else on the team earned $20. I can’t in good conscience promote you over them.”
The ones who held out until someone else took the hit, who earned the big bucks? They proved themselves. Their portfolio is full of big ticket items and significant business impact. They deserve to be celebrated. And in celebrating them for it, you reinforce their behavior. They will become even less inclined to volunteer in future.
Humans are great at learning from observing others, and if everyone can see not-volunteering breeding winners, then you will suddenly find yourself with rather a lot of not-volunteers.
This is how you lose your volunteers. This is how you build an accidentally selfish team.
If you’re very lucky, you may find yourself with a volunteer who persists even in the face of everyone else’s success. They may well be the ones who do the math and say “well, $1.25 is still better than $1, and no one else is going to press the button…”. But the longer they play this game, the more disengaged they will become. Ultimately, they will leave your team for more volunteer-rich environments.
This pattern doesn’t just play out down in the coal mines of day to day coding either. It also plays out in management. Indeed, it often plays out in a much more brutal way, as only those who have won at the game in the past get promoted into management positions.
Which of your managers warmly welcomes the brand new interns and juniors to their teams, even with their high ramp up cost, because they know that in a year or two those juniors will be contributing a lot to the whole company? When a talented new hire starts, which of your managers suggests that person might actually be more effective in someone else’s team? When a shared system needs a major overhaul, which of your managers volunteers their team to take it on, even knowing that said overhaul will not contribute towards any of their KPIs?
For managers though, the impact of volunteering doesn’t just affect their own success. It affects the success of every person who reports to them. Being a member of lowest scoring team in the company hurts every team member’s professional career.
The manager’s staff consequently become frustrated, pushing their manager even harder to stop volunteering. Eventually, the few volunteers who made it into management will crumble under the pressure from both sides, and stop volunteering as well.
With no one role modeling volunteering in management, those few volunteers who remain lose hope. Your accidentally selfish team becomes an accidentally selfish company, with everyone getting the bare minimum reward for just showing up, everyone starts blaming everyone else for not having pressed the button.
So what can we do to retain those volunteers?
At first glance, it seems like we should just adjust the metrics of success. Find ways to make sure that your volunteers are celebrated and rewarded as much, if not more so, than those who simply reap the rewards.
But that is hard. Really, really, hard. It is in the nature of the volunteer to take on tasks are required, but not rewarded. If you adjust the reward structure to value the tasks which are currently being volunteered for, you can be assured that something new will become a low reward task, and that task will require volunteers, and the problem will just rearrange itself to suit the new structure.
Changing which activities are rewarded won’t help, but systems do exist which reward intention over output. Some companies have managed to develop a secondary, parallel success structure based around a kind of “good citizen social capital”. This peer esteem can be an intrinsic reward, helping to change underlying dynamics in such a way that periodic volunteering becomes a minimum expectation rather than a burden to be avoided. These systems are easiest to create with new teams though, and I have not yet seen it successfully retrofitted into an existing team that is already set in its ways.
A slightly more complex answer is a ‘three musketeers’ strategy: all for one and one for all. If the team succeeds, then all people in that team succeed equally. If the team fails, then everyone fails together. Taking on an undesirable task to benefit the team should, theoretically, result in an equivalent ultimate reward for everyone.
I have seen this work in the past, but it requires a very strong culture of collaboration and teamwork to sustain it, along with a decent initial set of volunteers to get started with. In the absence of such a culture, those non-volunteers can quickly learn that success can still be won by doing absolutely nothing, and everyone suffers.
It also falls apart when the company inevitably asks who the one or two people are who should be singled out for promotion. It is a model which can be fundamentally at odds with many of our traditional systems of reward and remuneration, so tread with caution.
The third option is one which requires a very mindful, very patient, very present manager. When these thankless tasks come up, you as the manager of a team must watch who is volunteering. If you see it happening often, call it out. Something like “Thank you x, but I’ve seen you take on a lot of these tasks for the team over the past few weeks. I’d like someone else to take this one”. And if no one else will volunteer, voluntell. Require someone else to take the task.
This won’t fix the culture by itself, but it will give your volunteers a bit of space to build their own successes, and at the same time, remind them that their efforts have been seen, recognized and appreciated.
Volunteers are the real team builders in any organization. It’s so important that, as managers, we not alienate them, and wherever possible, try to build a culture where being a volunteer is something to aspire to, not something to be penalized for.
Maybe even aim for a culture where everyone is a volunteer, and we share the rewards of that equally.
