Social exclusion for bad apples

So far, we have learned

  1. The bad apple effect – a single, toxic team member can create group-wide dysfunction
  2. Emotions are contagious – when someone else experiences an emotion, mirror neurons light up the exact same areas of our brain as if we were experiencing that emotion ourselves.
  3. Social contracts are a nice way of establishing a set of desired group norms, and can provide a “container” for the team to manage its own conflict, including defective behaviour.

In this post, the forth in the Agile Team series, we look into techniques for working with people who continually act out of sync with the team culture and norms. In particular, we investigate how social exclusion can be used as a highly effective deterrent of bad behaviour.

The bad apple syndrome is a well-known phenomenon and there is a considerable body of evidence[i] proving that often group members strive to contribute at or slightly above the level of the person making the lowest contribution in the group – otherwise known as the Minimal Reciprocity Rule.

The key questions to conclude the Agile Team series are

  1. Why do some people consistently behave this way?
  2. What can be done about it?

To answer the first question – why – let’s look at some basic primal human motivation.

We all have conflicting needs and at times our personal welfare and the collective welfare of the group are in conflict. Sometimes, uncooperative behaviour is more immediately and personally rewarding than cooperative behaviour. In particular, when there is a culture of bad behaviour, we often protect ourselves first, and then look to the needs of the group.

This is completely logical; the limbic system (a primal part of the brain) quickly analyses each social situation as a potential threat or reward. If it is a threat the brain perceives, then the core flight or fight responses kick in and we go into “survival” mode. As a matter of fact, recent research suggests that these responses are very similar to ones that drive humans to food and away from predators. Naturally, if we sense a situation that doesn’t feel safe then we look after ourselves first. Of course there is a stark paradox here; by doing this we add to the culture of bad behaviour, thereby creating (and often justifying) more “bad” behaviour and a vicous cycle ensures. In contrast, the more cooperative others in the group are, the more cooperative we tend to be.


The challenge is nicely summarised in a fascinating article in the Journal of Experimental Social Psychology: “social dilemmas, by definition, present clear and often strong incentives to act uncooperatively. “. The article goes on to say “There appears to be a non-trivial fraction of the population that will nearly always defect. It would be surprising then if most groups did not include at least a few uncooperative ‘bad apples’.”

So in summary, there are strong incentives to revert to individualistic behaviour, a reasonable likelihood of there being a few “bad apples” in the group, and “bad” group behaviour initiates a vicious, self-reinforcing behavioural cycle.  I am sure we all have an occurrence of this etched in our minds – where we looked after ourselves first as there was little inspiration to work for the greater good of the group.

So what can we do about this? What do we do when working with a team that isn’t really working, primarily due to the poor behaviour of a few key people? And agile coaches, what can we give the team in order for them to resolve this situation themselves? And how do we avoid reverting to old, unsuccessful traditional management approaches?

My experience as an agile coach has taught me the following:

  1. Start with a focus on core values. However there is a catch here – values are not something written down in a document or put on a plaque on the wall. They are something that must be lived and demonstrated as behaviours by us all, starting with you.
  2. As a team, we need to have the desire to be better. If we don’t want to make a better working environment, then we won’t.
  3. We need to design what a safe, enjoyable and stimulating work environment feels like. We can’t expect it to just happen, as if by magic. Establishing a set of group norms, or a social contract, really helps.
  4. We need to be brave and open enough to confront one another when we perceive another team member’s behaviour as inconsistent with our agreed norms.

All of these must occur – that is, one of these on their own is seldom enough.  For example, there is little point in having a social contract if nobody is prepared to call out inconsistent behaviour.

However, I have always felt there was something else missing in my tool kit – as if the above wasn’t quite enough.  A Behaviour Psychologist I discussed this with raised something interesting – the ABC of human behaviour:

  1. Antecedents – systems and functions that try to guide the desired behaviour
  2. Behaviour – the behaviour itself
  3. Consequences – what happens if the behaviour isn’t acceptable.

The belief is that we can’t change a person, but we can influence the way they behave by shaping the environment they function within. That makes sense to me – managing the system in which the person functions, but the idea of “consequences” really made me squirm. It made we think of the untold hours I spent in the principal’s office at school,  or the horrible work environments I had worked in, full of awkward, problematic managers who unwittingly or dispassionately destroyed any semblance of culture, creativity or engagement.


I find parenting is a great way to learn good coaching skills. The easy thing to do as a parent is to offer brutal consequences for unacceptable behaviour. These have traditionally come in the form of smacking, yelling and in extreme cases the use of a weapon against the child (the cane, the strap). Every ounce of me rejects this notion as barbaric, completely ineffective and often counter-productive, teaching children that violence is acceptable. It’s not.


Our family approach is based on social exclusion. “We don’t hit each other in our family. You are going into time out to think about your behaviour and what you are going to do to change it.” By using this approach we are:

  1. rejecting them from the social group temporarily
  2. targeting the behaviour, not the person
  3. providing them an opportunity to take ownership of their behaviour.

It was uncanny how often I found myself saying “if only we could use this approach at work!”. Eventually, I found the time to research social exclusion as a workplace tool.

time out for adults

My research led me to an experiment conducted by Kerr, Rumble et al. They wanted to know whether social exclusion could prevent the viscous cycle of bad apple behavioural influence, and if so, under what conditions. They created two conditions for the experiment:

  1. Low Exclusion Threat-condition – where people operated with anonymity (no transparency) and there was no social sanctioning of any defectors (no courage, no respect, no commitment).
  2. High Exclusion Threat-condition, where their decisions were shared with the rest of the group transparently and where participants were vulnerable to exclusion by others in the group.

Interestingly, transparency alone was not sufficient to alter cooperative behaviour. Only when there was both transparency and consequences (social exclusion) did members change their behaviour.

The experiment was a game called “The Investment Game.”  Groups consisted of 5 players, and each had $5 which they had to allocate (in whole dollars) between two accounts, a Personal Account and a Group Account.  Contributions to their Personal Account were kept personally. For example, if you put all $5 into your personal account then you got $5. For the Group Account, the sum of all contributions were doubled, and then divided equally between the five members of the group. So if everyone put all $5 into the Group Account then they would each get $10 back – they would all better off they all put their money into the Group Account.

Groups were randomly shown a table of the choices of previous groups, with bad apples simulated by maximum defection – that is, bad apples had put all their money into their own personal account and nothing towards the group. The table shown to the groups indicated whether 0, 1, 2 or 3 members had allocated $0 to the Group Account. The results are fascinating:

number bad apples

From this study, we can see that a single ‘bad apple’ was enough to materially reduce levels of cooperation if there are few if any social costs for an act of defection (the blue Low Threat line). In other words, if conditions prevent the group from identifying who may be defecting and/or group members have no way of punishing defectors, cost-free defection is a possibility. Under such conditions, one bad apple can indeed spoil the barrel.

However, the results of the experiment also show that the relationship between the number of defecting bad apples and one’s own cooperation is rather different when both transparency and the threat of social exclusion are present.  When both are present, it took a majority of bad apples to produce a detectable decline in cooperation.

If following the example of a single or a pair of bad apples meant risking social exclusion, the temptation to follow that example was resisted. It was only when a majority prevailed that others crumbled to the temptation. Thus, they concluded that social exclusion can be an effective means of social control.


So this helped me immensely. It turns out social exclusion IS an effective tool to preventing and addressing bad behaviour.  Interestingly, another study I found helped me understand why.

Data gathered from MRI scans, suggests that social exclusion can result in a neural impulse similar and as powerful as physical pain.

“When people felt excluded we saw activity in the dorsal portion of the anterior cingulate cortex – the neural region involved in the distressing component, or what is sometimes referred to as the ‘suffering’ component of pain. Those people who felt the most rejected had the highest levels of activity in this region – Naomi Eisenberger, Social Neuroscience researcher, UCLA. Literally, social exclusion is broadly equivalent to physical pain.

Literally, social exclusion triggers the same parts of the brain as physical pain.

social pain

So, how can we use this information as agile coaches?

  1. Leadership – to me this is the first and foremost. Don’t expect others to behave in a positive manner if you can’t behave this way yourself. As my old friend Dave Acraman once said to me “when your kids are misbehaving, the first person to look at is yourself”.
  1. Give them the authority, autonomy and guidelines to self-manage and take ownership of their group behaviour. How would they like it to be? What would make them really powerful as a team? Help them write it down, discuss it, demonstrate it, notice it, reward it. A social contract is a good way of doing this, but make sure it is prominent (perhaps in the team area, or close to their Scrum board) and discussed regularly (retrospectives are a good place for this!)
  1. Teach and guide them to have the courage to detect unacceptable behaviour and be transparent about it. Sometimes an agreed approach can help the less confident of those in the team. I have used the NVC approach of “when you do X, it makes me feel like Y…”
  1. Help them learn how to have “consequences” for uncooperative behaviour via social exclusion. This may be as simple as “in our team we don’t treat other people like that” or a team member coaching a colleague on how to approach a situation differently.


I hope you have found the Agile Team series helpful. As always, your feedback is welcome.

Sorry it took my so long to complete this final post!


[i] Bad Apple Syndrome

  • Bornstein, G., & Ben-Yossef, M. (1994). Cooperation in intergroup and single-group social dilemmas. Journal of Experimental Social Psychology, 30, 52–67.
  • Braver, S. R., & Barnett, B. (1974). Perception of opponent’s motives and cooperation in a mixed-motive game. Journal of Conflict Resolution, 18, 686–699.
  • Dawes, R. M., McTavish, J., & Shaklee, H. (1977). Behavior, communication, and assumptions about other people’s behavior in a commons dilemma situation. Journal of Personality and Social Psychology, 35, 1–11.
  • Komorita, S. S., Parks, C. D., & Hulbert, L. G. (1992). Reciprocity and the induction of cooperation in social dilemmas. Journal of Personality and Social Psychology, 62, 607–617.
  • Messick, D. M., Wilke, H., Brewer, M. B., Kramer, R. M., Zemke, P. E., & Lui, L. (1983). Individual adaptations and structural change as solutions to social dilemmas. Journal of Personality and Social Psychology, 44, 294–309.
  • Schroeder, D. A., Jensen, T. D., Reed, A. J., Sullivan, D. K., & Schwab, M. (1983). The actions of others as determinants of behavior in social trap situations. Journal of Experimental Social Psychology, 19, 522–539.
  • Yamagishi, T. (1986). The provision of a sanctioning system as a public good. Journal of Personality and Social Psychology, 51, 110–116.


  1. Fascinating

  2. This was an interesting read for me thanks.

    One thing I’m not clear on is what ‘social exclusion’ might look like in practical terms within say an agile delivery team? Presumably I’m not going to get away with sending my stick-in-the-mud to the naughty corner!

    Can you give some concrete examples of how you’ve effected social exclusion within your teams?

    • Edwin Dando says:

      Hi Mike,

      Thanks for the feedback. In practical terms, here is how I tend to approach this.

      First off, I look at the overall environment. Most behaviour is a product of the environment. Is it safe? Is there trust? Do people care about each other? Or is it toxic, backstabbing, individualistic and hostile? The environment will always be a factor in the behaviour. Check out my article on agile culture here – it might help.

      I have facilitated sessions with the team where we openly discuss and visualise the ideal culture. It literally starts with a 10-minute mediation to get everyone’s heads clear, then a guided process (still with their eyes closed) to consider their perfect day at work:

      1. they visualise waking up – what does it feel like? What do you notice? Where are you? Who is there with you? What is your bedroom like? What is the time?
      2. Visualise what it is like to get ready for work. How do you feel as you brush their teeth and eat breakfast? Are you panicking and rushing? Are you excited? Are you anxious?
      3. Visualise the commute to work. How does it feel? What are you thinking? Are you calm? Are you scared? Are you excited about the challenge in front of you and being with your team? Why is that?
      4. What is it like as you walk through the front door at work? Are you relaxed, confident, inspired? Are you tight in the chest, scared and walking quickly? Is your heart racing? What does it feel like? What words describe this?
      5. What does it feel like to be at work? Are you watching the clock? Are you energised? Are are you just hoping the day will end?
      6. How does it feel explaining your day to your partner as you unwind at the end of the day?

      We then gently come out of that visualisation and try to come with words that describe it. As the facilitator, I just write any words that come to mind for them onto a whiteboard. Once we’ve captured enough I try to group it into some summary. I like mindmaps for this. From this we try to come up with some values we aspire to that represent our collective views. Not the typical corporate junk that nobody reads, but real sentences that describe real scenarios that explain their visualisation. It is critical to include examples, not just aspirational words. For example, in Clarus one of our values was integrity. We explained this with a number of scenarios, one being “it can be hard to know what to do when a client needs to hear the truth, but you are worried that telling them this might lose Clarus the engagement. We strongly believe in doing what is right for our clients and are prepared to put this ahead of revenue. Be honest, be true to yourself and them, be diplomatic, be caring, be brave and tell the client the truth. We will always support this approach.”

      So now we have some shared, co-created values and behaviours we aspire to. With that founding document in place, here is how I approach it.

      • I work as a servant leader to the team to help shape an environment that supports how they want to work. This might be simply getting them the things they need, or removing/resolving things in the organisation that are counter to their values.
      • Hold them to the agreement – let them manage the bad apple, not you. In my experience, the best way to change people’s behaviour is via their team/peers. Therefore, key function for you know is holding the team to account on their social contract. There is one thing to have a social contract, there is another to use it. As a coach, one of your approaches is to observe when you see a departure from their desired state and simply point it out. Don’t feel you need to resolve it for them. Your job is simply note it and see how they choose to respond to that. The third person approach works well – “I am picking up some tensions with this issue. Do I have that wrong or do you feel that too?”
      • Helping them process this tension: if they do feel they it too then you need to continue to remain impartial – “what would you like to do about that?”. Remember – your job is not to solve the problem. That is disempowering and prevents self-organisation. Your role now is to help them by being a true facilitator – let them all participate in how they would like to resolve this, and thus grow and improve.
      • Mentoring them: if they are genuinely stuck, and specifically ask for your help with words like “we are really stuck here Mike – can you help us? We don’t know what to do” then it might be time to nudge them along a little with some mentoring. Remember that you are changing stances now, from coach to mentor. Again, try not to give solutions, but teach them techniques and tools for solving these sorts of challenges. For example, Impact Feedback is a great one. Another approach is to tell stories about how other teams you’ve worked with have approached this sort of situation. If you don’t have examples then just make up a story. You don’t have to be disingenuous -simple tell a story – “there was a team once that ….” If they don’t want to solve the problem, then simply roll with this – they own the problem not you. Let them live with it for another week or two and when it comes up again rinse and repeat. Don’t solve it for them.
      • Once they do develop a plan for addressing the issue, as the facilitator/coach ensure you use a simple, quick way of capturing the decision and if you see them slip into bad behaviour again, bring them back to what they agreed.

      In parallel to the above, I ensure I establish a discrete, open dialogue with any required manager or HR department. It means that if things blow up then they are already across the situation and it won’t come as a surprise.

      I have used this approach with great success. My job is to hold a mirror to the team and let them decide whether and how they want to grow. I have use this approach to help teams confront bad apples and ask them to change. They have worked through seemingly impossible situations – blatant sexism, racism, abuse, micro-management, selfishness, individualism. The most extreme outcomes I have seen is the team asking a member to leave the team, or the person in question resigning as they quite honestly feel they are not capable or willing to change their behaviour.

      I hope this helps! If not, drop me an email at and we can swap numbers and do some remote coaching over the phone.

%d bloggers like this: