Made by
Made by
Jamie Cantrell
As a certified change manager, Jamie Cantrell understands the practicalities in behavioral change. In this session from DevRelCon San Francisco 2019, Jamie shares some key lessons from the process of overhauling community norms at SAP.
For those of you who were here in the last talk, we’re going to talk a little bit about those psychotic monkeys he was referencing when we talk about aggression, and bad behaviors, and what makes a community productive and healthy versus we’ve all seen those bad behaviors, right? The “Read the freakin manual,” and “Why don’t you just Google it?” So, that’s what we’re here to talk about today.
I’ve had a lot of conversations lately about the intersection of these two disciplines. Community management, for those who aren’t as familiar with it, is basically we take Developer Relations, and we have to scale it.
We are very concerned with the group dynamics, making sure that people are feeling like they are welcome to participate, that they have something valuable to contribute. So that’s what our talk is going to be about today.
Community leaders are essentially the conductor that runs the train. We make sure that the community gets where it needs to be on time, on the right track, and safely. And what that means is we’re primarily, in many cases, responsible for managing the platform.
We’re also responsible for making sure that people’s interactions with each other are safe and welcoming. So what you may have heard, again, if you were here in the last talk is talking about how to interact one-on-one with someone.
Community managers take that role, but they become, then… Thank you. They become essentially the mediators. We have to make sure that everybody is interacting with each other in a way, again, that makes them feel like they’re a valuable contributor to our community.
If you’re here to figure out if diversity is a good thing, that’s a different talk. We’re going to have to just agree that diversity is a good thing. So, when we get into talking about psychological safety, this is really what we’re talking about, making sure that we are having a diverse representation of opinions and experiences throughout our community.
We know that this isn’t happening when we see certain metrics, which we’ll get into a little bit later. But essentially, we want to make sure that people feel safe to take risks without fear of retribution or repercussions.
What this means is I should be able to walk into a room and say, “Hey, I have a great idea. Let’s do it this way.” Or, “I don’t understand. Can you explain it to me?” without someone telling me, “You’re an idiot. Why don’t you go figure it out for yourself? I’m not here to do your job for you.”
So, when we’re looking at a community, we are dealing with that dynamic that maybe many of you have seen in a team that you’ve been on, or anywhere on the internet. We have to make sure that we are facilitating those conversations.
There are many factors that go into somebody’s psychological safety, their sense of safety. One of the core ones that we see a lot, if you follow any of the diversity and inclusion conversations going on, is representation.
We have to see others who look like us in the community. It’s even better if we see them in positions of power and influence. So, as a community manager, if you are building influencer groups, you need to make sure that you are building groups that are representative of a wide diversity of experiences.
So the other items you see here, shared goals, shared values, these are things that, even not in a digital community, any community, any group that you belong to, typically you’re there because you have some sort of shared experience, shared objectives. You’re there because you can relate to other people.
Now, for us, we have a massive portfolio at SAP. And because of that, our community is, as Tamao referenced, 2.8 million members strong. What this also means is that we can’t have a community team that necessarily is knowledgeable about every one of those products.
So, our job is not to act as the direct moderators and product experts, our job is to empower those people who do know our products to then work with the beginners, the people who are just coming into our ecosystem and helping them to make sure they understand how to get started, how to develop their careers, how to become better SAP professionals.
We also have shared standards and expectations. So this kind of goes in with shared values. Basically, we should all have an understanding of how we expect other people to behave.
So, when you’re on the internet, you may have an expectation that, “Oh, trolls are a normal thing, whatever,” but we want to kill that expectation. We want to say, “No, we don’t accept that.” In our community, you won’t behave that way because we don’t expect that of you.
And the last one here is community enforcement. So, you will see this reflected later in my slides as well. Basically, this means empowering the community to self-police, so giving them the tools that they need when they do see bad behavior to say, “No, stop. We don’t like that. You can’t do that here.”
This is an important one when you’re talking about underrepresented groups. Risk tolerance is extremely personal. Depending upon what you’ve been through before you ever made it to our door, you may have very little emotional currency left.
So when you come in, we have to make sure that there’s as little resistance as possible to allow you to participate in our community. When you’re coming up against oppression, microaggressions, everyday things, by the time you get to us, you just want to do your job.
You don’t want to have to deal with bullies. You don’t want to have to deal with gatekeeping. So, our job is to make sure that that doesn’t happen.
I came up with this little cycle to show how the different ways, especially underrepresented groups, interact with a community that is not healthy, and it’s really, it feeds into itself.
So if you come into a community, you don’t see diverse voices represented, you don’t see people feeling like they can contribute, that they can speak up and say, “Hey, that’s not okay, or, “Hey, I have a different way to do this,” then you’re not going to want to participate.
You’re going to, A, feel pressured to act as the token person. So, if you’re the only woman in the room, suddenly, if there’s a question about, “Hey, what do women think about… Hey, Susie, what do women think about that?” And you’re like, “I know what I think about that. I don’t know what other black women think about that. I don’t know what other LGBTQ people feel about that.” It shouldn’t be on me to represent everyone who looks like me.
You’re also at risk of encountering more of that biased bad behavior when you are the only one in the room. Your psychological safety goes down. You no longer feel like you should be able to speak up. Your participation goes down. And here we are back again, now you’re not participating, so the next person who comes in the room who looks like you doesn’t see you either.
So you can see that this all feeds into itself. We have a lot of opportunities here, as people speaking in our developer communities, to break these cycles.
I did want to add a point in here. Diversity is not the same as inclusion. So I know those things get paired together a lot. When we talk about inclusion, it means you have to actually make the effort. Go out there, say, “Hey, we want you here. There’s a space for you here. Come join us.”
I also want to point out that the psychotic monkeys are not necessarily psychotic monkeys. Really, what we’re looking at is bad behaviors. And when we get into a place where people have been socialized to act a certain way, to believe certain things about different people when they have been living in internet culture forever, and they’re sitting anonymously behind their computers, these things add up.
And eventually, they start believing that certain bad behaviors are okay, that it’s acceptable to take out their emotions on others. This is what we have to address.
So, as community managers, again, we’re looking at people. We have whole people that have a lot of complex emotions and experiences that have developed into a series of bad behavior.
I mentioned we’d talk a little bit about metrics. So, there is a way for you to assess how healthy your community is today. We want to look at your demographic representation.
Given data privacy issues, and I was also very instrumental in implementing our GDPR work, so I’m very sensitive to this, you may or may not have demographic information, or you may have a limited amount. And, there are arguments to be had about the safety and ethics of gathering this information.
But if you do have it, take a look at it. See what it is that’s popping up that you’re going, “Hey, this is really strange. We have way a whole lot of these people and not very many of these people, but we know they’re out there in our community, holistically, our community, not necessarily on our website.”
So, we want to figure out, “Okay, is there a way for us to go out and invite those people in?”
Your stickiness. How often are people coming back once they’ve come to your site? As a new member, when I come to your site the first time, what kind of behavior am I seeing? Am I seeing people who are being really bad to each other, who are really bullying each other, they’re making it not a fun experience? I’m probably not going to come back, right? So, we want to look at that.
We want to look at engagement rates. So how many of your visitors are actually taking an action on your site versus just coming and taking a look?
And then member satisfaction, this is something that you basically have to survey and talk to people. You have to go out and make the effort to find out, what is your experience like with our community?
There’s lots of others you can do. This is a good starting set.
So, one other thing you should know about me is I am also a certified change manager. So, my focus in a lot of this is human behavior and looking at how do we change people’s attitudes? How do we change habits?
Because a lot of the bad behaviors that we see on the internet are bad habits. When your first response is to go, “LOL, noob,” at somebody because they’ve asked some question that you’ve heard 100 times, you would never do that in person, right? But we’ve become so numb to this, and we’ve developed these bad habits that we have to take now a proactive approach to changing those behaviors.
The first step is to develop your code of conduct. So, if you don’t have a code of conduct, you need to write one.
If you have one, and this is an exercise we actually went through within the last year or two, you need to actually sit down and look at it and say, “Okay, am I creating loopholes with this by trying to make my code of conduct very specific to particular incidents we’ve dealt with in the past?” If so, you’re going to have others who go, “Ahh, I didn’t technically break the rules, so you can’t do anything to me.”
Instead, we need to make sure that the Code of Conduct is broad enough that it covers general bad behavior. We’ve seen this in a couple of major community sites like Stack Overflow, where, you know, they’ve taken an approach to try to fix the culture, but we have to make sure that we are making it broad enough that it’s enforceable, it’s clear, and understandable by everybody. Again, that goes back to that shared values and expectations point.
It’s concise, we don’t want something that people are going to look at and go, “Pfff, I’m not going to read this. What is this? The terms and conditions? Like, no.”
And, again, it has to reflect our shared values and goals. It needs to be something that everyone can buy into and say, “Yeah, I agree with this, and we should enforce this.”
And lastly, it must have clear consequences. One of the things we’ll talk about in a little bit is consequences. We have to be transparent and make sure that people understand what action we’re going to take. As the caretakers of their community, as the conductor of their train, what are we going to do to course correct?
So the first thing we need to talk about when we’re looking at adjusting behaviors is what kinds of things motivate your community members? Motivation is, again, extremely personal.
It will be a combination of intrinsic, which is very internal. These are things that are generated from within you. So, if you are working on a project because you’re really excited about it, but you’re not getting paid, you’re not going to show it at work, you know, this is just something you like to do, that’s an intrinsic motivation.
There are also extrinsic motivators. These are much more clear, obvious motivations. So, if you’re getting paid to do your job, that’s why you go to work. If you are excited to go to your conference because you’re getting swag. Gamification is something we use quite a lot in community management for this purpose.
Career advancement opportunities, these are all things that are external, and they don’t necessarily have to be positive exciting things either, right? These could be obligations, like, “I have work obligations. If I don’t do my project, I’m probably going to get fired.” So, anything that is an external force upon you.
So, we have to understand, for our users, generally, how can we motivate them to make the behavioral changes we want them to make?
These are some good questions to ask and to think about what your users are asking themselves as you’re developing your plan to make this cultural change in your community.
What makes me feel good? What do I hear from others who I respect? What happens if I don’t change? Are there consequences to that?
And, finally, what happens if we as a community don’t change? What are the consequences to others within our community or potential members of our community if we don’t make this change? This is where your influencers come in.
So, theoretically, every community has a couple of different tiers of contributors of members who participate in different ways.
So you’re going to have, the broadest level is going to be your lurkers. These are people who just consume content. They’re not really there to generate stuff. They might be looking through and finding answers to their questions and moving on with their day.
The smallest group and the most powerful group that you have to get on your side is your influencers.
So when we talk about a community like Stack Overflow that has had some cultural issues in the past, one of the things that I’ve personally observed and think that they have an opportunity in is utilizing influencers more to drive their message.
We have a couple of influencer groups within our community. We have also gone through and actually culled some of the detractors from our influencer groups. You have to be very wise about who you’re choosing to carry your message forward.
You want to identify the people who are going to be your champions. They don’t have to follow without question, but they, ultimately, at the end of the day, have to be there for the benefit of the community.
This is also when we get into that self-policing, right?
So if you have influencers out there, you have your high-level moderators, your people who have a bajillion gamification points, or however it is that you decide to organize and rank your community, if those people are out there saying, “No, you don’t respond to somebody like that. We don’t treat people like that. We don’t do that here,” then all of that trickles down, and the rest of your members start to fall in line, and they go, “Oh, that’s right. We wrote that code of conduct. I guess we’re actually doing that.”
So, this is really, really a critical component of making this change for your community.
Now, we’ll get a little bit into your user interface. Many of you, if you have a community, are probably using a vendor solution of some kind, whether it’s Lithium, or I heard Discourse earlier.
We actually have our own combination of AnswerHub and WordPress implemented, so we have a lot of control over what UI elements make it into our community. These are really critical for the self-policing aspect.
So, for the community to actually enforce what is happening, what they’re seeing, and, again, there’s six of us on our team, and there’s 2.8 million users, so there’s a lot of activity that we don’t see, we have to give them the tools.
It starts with what we can control, right? So, if we have a list of terms that we know is just being used to hurt and torture people, we can certainly set that up as a blacklist and say, “Ah-ah. We don’t use those.”
But we can also give them several different options for controlling their experience and for setting expectations within the community.
So, reporting this content is a big deal. You need to be able to say, “Hey, I found this thing. Somebody needs to take a look at it. I feel like it doesn’t really align with our values. It doesn’t really match our code of conduct.”
Muting and blocking this user gives somebody personal control over their own experience. And I skipped one on purpose, I’ll come back to that one.
“Report this user” allows the community as a whole to come together and say, “Hey, this person is really not acting in a way that’s beneficial to our community. We need to do something about that.” So, that’s on your moderation. You know, what do you do with that?
The one I skipped is data privacy and control, and I mentioned that I have worked on our GDPR implementation, which had both user interface and policy ramifications, obviously.
I really believe firmly that users should have as much control as we can possibly give them over their own data. So, that means that we have to allow them to leave. Any good relationship, you should be allowed to leave.
So, this means giving them as-easy-as-possible ability to delete their data, to hide their data. So, for example, we have the ability for our users to hide their profile from being public as often as they want. They can turn it on and off.
It’s kind of annoying for other users because you may have just turned it on to answer a question and then turn it back off again. But to us, it was more important for our users to have that control over their data, over their exposure to the community and to other users. And, again, these are safety issues, in the end.
And lastly, so I briefly mentioned moderation policies, and I mentioned earlier also transparency. So, to tie this all together, you really have to show how you’re going to enforce these things, right?
You have a Code of Conduct, here’s what we’re going to do about it if you don’t follow it.
And one of the things that I’ve learned, actually, from other community professionals is the value of reporting on what you’ve actually done. So, what issues have been reported? What did we do about it? And it doesn’t have to be, you know, “So and so said this about someone,” but more, “We’ve had five issues with bullying, here’s the actions that we took.”
So, in summary, psychological safety is really critical to building a diverse and inclusive community. Cultural change is a process. You will lose people along the way. There will be people who are unwilling or unable to change, and that’s okay. In the end, you’re going to be left with the people who are there for the benefit of your other members.
Choose your influencers wisely, they will carry forward your message, or they will destroy it. And lastly, risk tolerance is deeply personal. So, be aware of that as you are building out your policy, as you’re building out your change efforts, and ultimately, you will hopefully come out with a safer community. Thank you.