This is a transcript of episode 148 of the Troubleshooting Agile podcast with Jeffrey Fredrick and Douglas Squirrel.

A listener asks us to explain the link between “normalisation of deviance” and fear. Using examples like an agile team dropping its retrospectives, or NASA launching the Space Shuttle in too-cold conditions, we illustrate how being afraid can drive a group away from its espoused norms and toward dangerous alternatives, and conversely how you can use examples of “normalised deviance” to find and mitigate hidden fears.

Show links:

Listen to the episode on SoundCloud or Apple Podcasts.


Listen to this section at 00:14

Squirrel: Welcome back to Troubleshooting Agile. Hi there, Jeffrey.

Jeffrey: Hi Squirrel.

Squirrel: So we have a listener question today, I think.

Jeffrey: That’s right. Before we get to it, though, I just want to share with our listeners something I’m very excited about, which is that I was a guest on Gene Kim’s Ideal Cast, which was published this week, and we’ll put a link in the show notes. And so if you’ve thought, you know, Jeffrey, enjoy your podcast, but we’d like to hear even more then you can check that out. I wil say one thing, it’s a very different format, which is that it’s an almost two hour conversation with Gene Kim but it was a lot of fun to do. And I hope our listeners check it out and enjoy that as well.

Squirrel: Fantastic. I know I plan to. Excellent. Well, let’s get onto that listener question then. Jeffrey. What did our listener ask us?

Jeffrey: Yep. This is a listener and actually a member of our Slack community. And so if people think they’d like to ask us questions and have interactions with other people who have listened to the podcast and have read the book, Agile Conversations, then go ahead and get in touch and we can invite you to our channel because we got this via Slack. And this is from Horia. And he says, I’ve been re-reading Chapter four and this is of Agile Conversations, which is the Fear Conversation chapter. “I’ve been re-reading Chapter four and this quote, keeps bothering me to ‘prepare for the fear conversation, look for examples of a dangerous deviant norm in our organisation’. The question for me is how does analysing existing normalised deviance help us prepare for the fear conversation?”

Jeffrey: And you jumped right in on Slack with an answer Squirrel but I thought we could use the podcast this week to give a little bit more expanded version of that. And for people who haven’t read the book, maybe you could explain a bit what we cover in Chapter four here, what’s the fear conversation about and how does normalisation of deviance help us discuss it?

Squirrel: I’ll do my best, but you’re going to have to help me, Jeffrey. OK, so let’s cover what some of the words are. So ‘fear’ I hope most people know what fear is. So I’m not going to give a big definition of it, but it is one of the major obstacles to a successful Agile transformation or DevOps movement or anything else that you might want to do. If your team is afraid, they are unlikely to take risks and try new stuff, that’s not terribly revolutionary to say. The problem is what on earth do you do about fear? So we give several steps that are helpful and we’re going to focus on one of those. The others are things like, for example, coherence busting and improving conversational skills, which we cover in other parts of the book. So we won’t talk too much about those. But once you’ve improved your conversational skills, once you’ve got to the stage where you’re able to approach the problem with curiosity, then to say, ‘gosh, we sure do seem to have some fear here. Where’s it coming from?’ That’s where normalisation of deviance can come in and help you. And what is normalisation of deviance? Well, the classic example from the book by I hope I’m getting her name right, Diane Vaughan on the Challenger Disaster.

NASA’s Mistake

Listen to this section at 03:12

Squirrel: That’s the source that kind of defined the term, although it’s used in lots of other places. The example there is the space shuttle Challenger and some of our listeners will be too young to remember it. And I certainly remember being in school and hearing about this explosion, which had a teacher on board. They were starting to send civilians into space. And the analysis afterward revealed a lot of difficulties in the assessment of flight risk and in particular, the assessment of the risk of flying the space shuttle on a cold morning when the temperature is at or below freezing. In fact, they’re sort of, in retrospect, terrifying photographs of icicles forming on various objects near the launch pad. And it turns out that certain elements of the space shuttle didn’t deal very well with freezing temperatures and they expanded when they shouldn’t and they shrank when they shouldn’t. And that led to an explosion.

Squirrel: How did they get there? What’s this got to do with normalisation of deviance? Well, the engineers knew, the engineers were aware that there were significant risks to flying the space shuttle under certain conditions, including very cold ones. And they had been shouting ‘the risks here are 1 in 10, 1 in 100, you know, very high risk of something going wrong in.’ The managers had a norm which didn’t match, which was get the space shuttle in the air, don’t delay the launch. And by adhering to that norm instead of the norm that the agency asserted that it followed, which was astronaut safety, first, they put astronauts at risk and in fact, caused the explosion. So that’s what normalisation of deviance is.

Squirrel: You get a different norm that the organisation begins to follow a sequence of steps, a set of actions, a pattern of action which does not match what it says it does. And the organisation gets comfortable with that. Everybody knows Oh yeh, well, we don’t follow that. “You know, our stand ups, we don’t really have them, we have them when we need them.” Retrospectives? “Yeah. You know, we had one, I think, a couple of months ago. Yeah. That’s something that we do when we need it.” Tests? “Yeah. We really think tests are important. I haven’t got so many. We’re really going to add some soon.” Those would all be examples of normalisation of deviance and they are signals of fear.

Jeffrey: They’re signals of fear. That’s really interesting because I think people might be surprised by that because I think many of our listeners will recognise the kind of statements you described sort of like, “well, you know, I know we’re supposed to do retrospectives, but we haven’t had one for a while.” I know that’s one I’ve seen many times. What’s the fear? You know, can you give me an example of the kind of fear that might be behind that?

Jeffrey: Absolutely. And I can give a couple of examples. So on that one specifically, there could be many fears. And this is the point that and this is answering Horia’s question, why is this thing ‘normalisation of deviance’ in the fear chapter? It’s because it’s a signal. It’s a clue. It’s a breadcrumb trail that can lead you to fears which might help you. So if you have the sense there’s fear and you have the sense there’s normalisation of deviance, but you haven’t got it all hooked up together, figure out what fear is underlying a specific normalisation of deviance. So in this example of retrospectives, you might be afraid of wasting time. You might be afraid of discovering something, making people uncomfortable, because they’d have to discuss an uncomfortable topic. “That Squirrel guy, he’s not very good at tests. And that’s why we haven’t written very many”. Confronting that might be challenging, might lead to a difficult conversation with that guy Squirrel. Might have to teach him how to write tests because he doesn’t know. So those are the kinds of fears that you might have in the retrospective example.

Squirrel: I’ve got a client who has this kind of situation and is working hard to mitigate the fears they have up on the wall, various customer themed statements about what they believe in, what is important to them, what they want to do. So they want their application to be fun. They want it to be family friendly. They want it to be easy and simple and all kinds of very natural, very nice, good things to say. In fact, if you go and look at their Jira board, what they have on there is a whole bunch of tasks that have to do with making their lives easier and making cost savings. So things like needing fewer people in the call centre and requiring fewer resources in order to offer their products, making their partners happy. Those are the things they’re actually working on. And when you go ask somebody, they say, “well, you know, if we didn’t do this stuff for these partners, they would fire us and then we wouldn’t be able to trade. We’d lose money.” And so for them, it’s the fear of losing money. It’s the fear of trading challenges. Very rational fear, by the way, in the current circumstances, with a pandemic on, lots of people are concerned about losing money so their fear isn’t wrong or misplaced. The problem is that they continue to espouse the original norm, the original belief we make things that are wonderful for customers, fun, easy for kids, all kinds of stuff, but they actually act in a completely different way. And it’s that disconnect that’s the problem. You could mitigated in many ways, including moving toward the norm that you’re actually following rather than the one you espouse. But continuing to hold both is where the danger comes, and that’s where the fear, conversation and mitigation of fear is helpful.

Learning to Communicate on an Oil Rig

Listen to this section at 08:35

Jeffrey: I really enjoyed you bringing this up, and it’s funny what it really brought to mind for me was an article I read many years ago that talked about safety on oil rigs. It was talking about Shell and that they had put in place a programme where they would teach people on how to talk about their fears because that had been a barrier to safety, then actually by putting this in place and having to learn to talk about their fears, they had improved their safety record by I think it was something like 84%. We’ll put a link in the show notes to the National Public Radio show where I came across this and it was to me, it’s a very similar thing, which is that that was an example where people had a culture of, like, you’re supposed to be tough. You’re supposed to work around these things. You know, what was wrong? How come you can’t man up? And the result was that people died. They had terrible accents, they were in a very literally dangerous environment. And the unwillingness to discuss their fears had these consequences. And what I like when we brought in this chapter then is being able to spot this gap between what we say we’re going to do and what we actually do and then tracing it back from our unwillingness either to be motivated in some consequence, but even just the idea of the fear of raising an issue, of being the only one to raise something and therefore standing out from everyone else. That certainly occurred to me as a direct application of this is that the being able to identify what was different and then using that and this is I think this is the key element in the chapter, is using that gap as a way to have the conversation.

If the Norms Don’t Fit, Change Them

Listen to this section at 10:43

Squirrel: Indeed. And then we go on, of course, to give specific steps and advice about how to have the fear conversation about tools you can use to get your mind set right. That’s coherence busting for example, amongst others. And we talk extensively about what actually you can do because we hate it when business books tell you this is the problem and then don’t tell you how to solve it. So we do work hard to make sure that there’s some tools for that. That’s not our focus today. That’s in the chapter if you want to read it or in various of our writings. But the focus today is on how do you even find out what’s going on? And we’ve kind of illustrated here for Horia, I hope, how the normalisation of deviance can help you kind of on the way in when you’re starting to have the fear conversation and you’re thinking, gosh, there’s some fear around here. I don’t know where it is or there’s this normalisation of deviance. I don’t know what it’s coming from. Either way, the one can help you to get to the other, but it’s also helpful on the way out. So once you’ve had the fear conversation, you’ve identified some things you can do to mitigate your fear.

Squirrel: “Hey, we’re afraid of wasting time. So we’re going to get a clear message from the top brass about what we should be investing in long term improvements, such as retrospectives.” That might be our mitigation step. And then we could describe what our new norm will be, what will be the restored norm, if you like. What will we normalise toward that isn’t deviant. And in that case, it could be, for example, ensuring that we invest X percent of our energy in long term improvements, which include retrospectives. And we’re going to get the top brass to help us understand what X should be. That would be just one example of something that you could then even print out and put up on your wall to accompany whatever existing normal, like being kid friendly or whatever it is that’s on your wall already there but you’re ignoring. You could say these are the fears that sometimes hold us back from expressing the other norms on this wall and here is how we would recognise that we are acting in accordance with the espoused norms and what we say we’re doing. For example, not necessarily we hold a retrospective every time, but that we hold a retrospective in line with the expectations of the management of the company about long term investments.

Jeffrey: This point of having established norms, of explicit norms as part of the outcome of a change is something I really like because I really like the way that it empowers people to then hold each other accountable for that element. They’re going to do, in fact, to relate this back to the Gene Kim podcast. One element in there that we were discussing, I had a bit of a discussion of when I was learning extreme programming and was first on a team that was doing pair programming. And I was having a conversation with my pair and we got to a point where I said, “well, what should we do now?” And what he said is, “you know what we could do, and I’m not saying we should, but if we were brave, what we would do now is write a test.” We were very early in learning TDD (test driven development) and working out those practises amongst ourselves and that idea of saying he could hold that up and say, “remember, this is what we signed up for. I’m not saying we should do it, but this is what we said we were going to do.”

Jeffrey: And that really helped us to take that next step and say, yeah, actually, that is what we said we were going to do. It’s it’s not the easiest thing for us to do right now. It’s not the most natural thing because we haven’t yet built those TTD muscles that writing a test now for me would seem like the obvious next step. But at the time I wasn’t yet there. But holding up that norm, “Remember, we said this is what we’re going to do” allowed us to follow through and do what we said we’re going to do because he could hold it up as the example. So it was a great example of this kind of principle you’re describing of having not just the mitigation, but then also the norm where the mitigation is the embodiment of it that you can use to hold up to remind yourself for what you said you’re going to do. I really like that.

Squirrel: Excellent. Well, Horia, I hope that’s been helpful, that you can use normalisation of deviance as a way to find fears and to identify things that you might want to mitigate in any sort of your fear portfolio. You’re always going to have fears. But which ones are you mitigating? Which ones are discussable? Normalisation of deviance, looking for it can help you find those or the fears can help you to understand the normalisation of deviance you know about. And it’s helpful for setting a norm and saying this is where we’re going to go to. That is not deviant. So that exactly as Jeffrey, you can be brave and say “we’re going to hold that retrospective, we’re going to write that test.” And having those explicit and discussable often allows the team to overcome their fears and act differently. Sounds fantastic.

Jeffrey: That’s great. I’m really happy we could give an extended answer on that and I hope that everyone enjoyed this. And any more questions? We’d love to hear from you either on Slack or Twitter or LinkedIn or email and any other place you can get a hold of us.

Squirrel: And you can get all of those things on or They all go to the same place so if you can’t remember, just find us and go to our websites and you’ll find all this material and ways to get in touch with us, which we really like. Like when Horia did that for us to give us a topic for today. And of course, if you hit the subscribe button, we’ll be back next week and you can hear us again. Your app of choice will tell you that we’re here and we’ll be talking to you again about troubleshooting Agile.

Squirrel: Thanks, Jeffrey.

Jeffrey: Thanks Squirrel.