Tech Leader Chats: How to build a resilient engineering culture with Meri Williams

Title of talk with photo of speaker

Culture, as Peter Drucker famously said, “eats strategy for breakfast.” The benefits of a strong culture can include:

  • Facilitating communication and collaboration
  • Motivating people to increase productivity
  • Mitigating burnout even in times of high stress

However, trying to develop or manage a good culture is difficult because, first, what is “good”? The culture that works in one organization may completely fail in another. Second, culture is as fluid, unbounded, and intangible as air. No one “owns” it and it’s constantly evolving.

Faced with such ambiguity, how are we as engineering leaders to act? How can we ensure that our actions are contributing to a “good” culture for our teams, the engineering org, and the company as a whole? If culture change is inevitable, how do we lead our teams and ourselves to adapt to such changes?

We’ll talk specifically about questions like:

  • What are the pros/cons of culture-fit versus culture-add?
  • How do you guard against hiring someone that is actually bad for culture? How might you respond if you realize this after they’ve joined and are having a negative impact on the team or organization?
  • How do you build a culture that accommodates a diversity of life experiences, world perspectives, and work styles, especially where they seem to be polar opposites?

About the speaker
Meri Williams is an experienced CTO (currently Pleo, formerly Healx, Monzo Bank, MOO, Marks & Spencer online) known for building and scaling technology organizations across a wide range of industries, countries, and sizes from small startups to companies with thousands of people. Meri is an author, international speaker, and Chair of The Lead Dev Conference. They are passionate about helping develop technical leaders.

How to build a resilient engineering culture

See below for:

  • Key takeaways from the talk
  • The recording
  • Resources noted in the talk
  • A transcript of the talk
Key takeaways
5 things to remember to build a resilient engineering culture

People need to hear things more than once

Give people purpose, autonomy, mastery, and inclusion

Watch for inflection points when challenges in the business change

Not everything can be tested, but it can be observed

Focus on getting the most out of people's differences

Recording and slides

You can view the slides from the talk here - and see below for the full recording



James Dong  0:00  
Great. So I'll say so Meri, for me is amazing for their work in doing the lead dev conferences. I just came back from lead dev west coast a few weeks ago and was absolutely so thrilled and impressed with how intentionally inclusive and diverse of the event was. So that was really wonderful. And then Meri is also really passionate about developing tech leaders as an example of that they are currently doing this talk at 10pm at night in their local time. So I'm really happy and excited to have him here. The one logistical note that I'll point out real quick is because we this is a big topic, we know. And we got a lot of questions in the background. So we actually the original plan on the agenda was to do 30 minutes of the talk and q&a and 30 minutes of breakout sessions. And because of the volume of people in questions, we're actually just going to take the time from the breakout sessions to do more q&a. So hopefully, that will give people more a chance to ask Meri, all of your big questions. And if you have any questions about that logistically for me, feel free to go ahead and send me a message. And then throughout the talk, if you have questions, feel free to drop them in the chat. Right, great. And then otherwise, Meri, I will let you start sharing your slides. Awesome.

Meri Williams  1:28  
Can everybody see? Okay? Awesome. So Hi, I'm Meri. And that intro is much nicer than the one I usually do about myself. So I won't list off all the all the things that I've done. Although the one the one thing I'd say is like, although I talk a lot these days about about people related topics, I really started out as a bit of a hardware hacker I my weird claim to fame is I built part of South Africans for satellite as a teenager. So my parenting advice for everybody is don't get let your kids do anything cool when they're young, because it's really hard to compete with the gaki 16 year old version of myself that sold or something that went into space. It's all been fucking downhill since then. So. So that's my, that's my parenting advice for all of you. But today, I'm going to talk about five things I've learned while scaling teams and scaling cultures, about how to make cultures like both healthy and resilient. And the first lesson that I've learned pretty much the hard way is that don't repeat yourself doesn't work for human communication. It's a great programming principle Umbra Python isa by background. And so it's one close to my heart, but it's terrible for human communication, we find that we need to repeat important things consistently, like seven times, you got to say it the different ways in different mediums, different times in order for people to really get it. And there's all sorts of implications that people don't know what's going on. Right? It not least among which that the only thing developers produce more often code is conspiracy theories in the absence of sufficient information. And so be clear and consistent. And remember that others still need to keep here beyond the point when you're worried that you're being repetitive. If you're worried that you're saying stuff over and over again, then you absolutely are onto almost the right point. And you just need to say it a few more times. The other thing that I think is important here, I love this, this this quote from David Morrison that the standard you walk past is the standard you accept, and I and so I think part of communication as well is calling out what is and isn't okay, and being really consistent about doing that. And we'll talk a little more in the q&a I'm sure about about what I mean by this. Remember, though, that we don't just communicate in the present. From a technical perspective, something like an architectural decision record is a really good way to communicate into the future, what you were thinking when you made a choice. And that having been on many, many, many, many legacy rescue missions, I worked on systems older than myself, I've worked on systems sold, they were not just legacy, but vintage, right? You desperately need to know what people were thinking when they made early choices. And so remember to communicate into the future in that way as well. I think that architectural decision records are great from a technical perspective. But if you make like fundamental choices about team day, you know, Team setup or organisation or the way that you do things as well, it can be super helpful to record your your reasons why so that people understand the intent behind it in the future as well. The second thing that I've learned over the years is that high performing teams needs you to create the conditions for success more than anything else. We all hate bad bosses. You know, we describe them as clueless, they're empty suits. They're pointless they, they practice the seagull old style of management fly in shouted everybody shit on everything fly away again, right? When none of us want to be a bad boss. And I remember when I, I kind of became a manager because it was the only way to move up in the world. When I was in my formative years, there wasn't a route stay an individual contributor. Honestly, if there had been, I probably would have stayed in icy for a hell of a lot longer if I could have done. But I remember having to become a manager and going well, if I have to do this, then I just flatly refuse to be terrible in that it like fuck that thing. In particular, I'm not going to be awful at this. And so I went looking for advice on what decent management looked like. I love this from Caterina Fake, you know, there's three types of managers the shit umbrella tries to protect their team, the ship funnel just picks the person they least like and concentrates all on to them ship and spreads it around indiscriminately. And I was like, okay, minimum viable, decent management is being a shit umbrella, I actually prefer an analogy that Nick means uses for this, which is to be a heat shield. Because with the heat, you know, you want, you end up with a shit umbrella and algae starting to talk about needing to be a little like semi transparent. So people still know what and that's just not a great image, right? You don't want to get into the, the imagery there. Whereas the heat shield is, you know that there's heat over there, but you're protected from it. I think that's a probably a more positive interpretation. And so I went looking for for research, because I'm a nerd. And because I'm a nerd, this is my favourite management book, you could all refuse to be seen in public with me afterwards, if it's too nerdy for you, but it's called first break all the rules. And it's my favourite because it's immensely databased. It's, they interviewed like 80,000 managers in hundreds of companies, and they were not looking for what made people happy. They were looking for what made people high performing. So they were like, Why is this restaurant so much better than the others in the chain? Why is this division of this company so much more profitable than the others? Why is this factory so much safer than the others every every sort of aspect of performance? They were they were looking at. And they discovered these 12 questions that predict high performance. And I know that this is both like the worst design slide in the world, and also impossible to absorb. So let's, rather than drowning in it, let's zoom out a little bit. I imagine many of you have watched or read Dan Pink's drive, where he talks about how, you know, the modern understanding in psychology of motivation says that we need three things we need purpose, autonomy, and mastery. So we need to believe in why we need to get a say in what and we need to be proud of how, and then you take away any negative factors that detract so if you have, you know, the worst commute in the world, maybe that won't be made up for by having great purpose, autonomy and mastery. But in general, those are the three things that people are looking for. And those 12 questions that I showed you a second ago, very briefly fit very well into that framework. Right. So one of them is does the mission or purpose of my company make me feel like my work is important? Do I know what's expected of me at work? Do my opinions seem to counter very clearly autonomy questions. But what surprised me at first was how many of these are about mastery. And then I thought about engineers and designers and product people I knew who had quit. And I was like, either it was a really bad boss, or it was mastery. And that actually rang true to me that nobody likes to be made to do poor work, or to feel like they're not growing in their craft. And I think some of these questions are a little old fashioned and how they're phrased, I think the Do I have the materials and equipment I need to do my work? Right question might have, you know, people and knowledge and those kinds of things in it if they wrote it today. But I still think it's a great question at work, do I have the opportunity to do what I do best every day is my favourite of all 12 of these questions. So if someone to work cares about my development, or my co workers committed to doing quality work, in the last six months, if I talk to someone about my development, and at work if I had opportunities to learn and grow, those are all about whether you feel like you're getting better at the how of your job, and you're learning and growing. But there's a few missing, right. There's three more questions in the last seven days have I received recognition or praise for good work? And it's super interesting that it's in the last week, right? Does my supervisor or someone at work seem to care about me as a person? And then do I have a best friend at work? This last one I find really funny, because in Northern Europe, where I give this talk a little too frequently, they hate this question. They're like the company does not get to decide who my best friend can be. Fuck you. And so I have to translate this in Northern Europe and possibly for the Australians and New Zealanders in the room as well. That this is like Is there somebody at the company who even if a company is not paying, you will willingly share a drink with that's the that's the definition to run with. Right. So it's a it's a masterclass in cultural differences. But I think this is about At California, he talks about this as community, I talked about it as inclusion, I think this is about feeling respected and rewarded, and that you can be yourself and succeed. And so if we think about these predictors of high performance, they all fit if you if you include inclusion, right, so if you purpose autonomy, mastery plus inclusion, turns into space in which you can be brilliant. So getting a say in y, sorry, believing in y getting a say, in what being proud of how and feeling like you belong, are the things that add up to a environment in which you can be wildly successful. And I have a personal belief, I suppose that every role and every person is capable of virtuosity, I, I don't think that there is a job that is not worth doing well. Nobody gets up in the morning thinking, today, I'm aiming for just south of mediocre and if I could fuck everybody else's day up along the way, my day would be perfect. I just don't think we're wired like that. Maybe the occasional true asshole does exist in the world. And you should find them and exit them as quickly as you can. But in general, we want to be great at what we do. And so our job as leaders and managers is to just get out of the way of that and let people be as brilliant as they want to be. The third thing I've learned is that there are real inflection points when you're growing teams and when you're growing your culture, sample and excrete them set outwards, aren't means always in the ship, just the depth varies. But in Latin.

And I think the the key learning here is the different things come for free at different inflection points. So when you're at at, you know, at an early point, you can see everything that's going on in the organisation, you can figure out when it's time for you to move up into a different role, because everything is just immediately obvious to you, you're 10 people or 20 people, you may not even need roles at that point in the company, right. But the important thing here is to focus on the right problems at the right time, I see so many people trying to take Google or Amazon or Apple or, you know, the Netflix are similar as huge company practices and try to make the work for themselves when they're in a fundamentally different different phase and different stage. And so like I said that, you know, 10 people, you don't need anything, you've got 10 people, they all know what's going on, everybody feels like they're fully utilised, because if there's anybody who's a little bit bored, it's immediately obvious, and you just give them more more stuff to do, you grab them and you or they grab themselves and add more to their plate right? At 50 people, you start to need a little bit of structure, you need to articulate what it looks like to progress, people start wishing for things like career frameworks, if they if they don't exist yet, at 100 People 150 People is what's known as Dunbar's number, the the science behind it is, maybe could be a bit a little bit more rigorous, but there's definitely some number beyond which we struggle to maintain so many human connections, right. And so villages traditionally were around 150 people, because that's the number of people you can actively know, without struggling too much, right. And so when you get to having groups bigger than 150, the US and then problems start to be very real, because that's more than we're wired to be able to cope with. And so that's why even in huge companies, you often see them divided down into around 100 person groups, because that tends to be to be what works quite well. The fourth thing I've learned is that with people observability matters more than testing, because however well we plan, however hard we try, our impact doesn't always equal our intent. And it's really, really tough to test things that are people related. You can try, you can trial, a do reviews process, you can trial, a different way of giving feedback with a small group. But if you'd want to restructure your organisation, it's very difficult to do that with a test group. And then and then do it with the whole group. Right, some changes you just have to make. And so regularly check in whether what you intended matches up to the actual impact. retros are a fantastic way to do this. And these are links to some some other talks that are this is from Jesse link, who's an amazing leader. She's VP of Engineering at Twitter. She's now senior director at Google. And it's a tour of different retrospective formats, what they're good for and what they're what they're great at. And, and I highly recommend it. I'll happily share the slides afterwards so people can grab the links. But also things like the Spotify squat health check model that's all about checking whether your changes are having the impact that you want to that you want to have and whether people are happy according to a set of metrics that they find useful. So don't just take Spotify as list from down the left here, develop your own with your team and then and then use the rack rating over time to to figure out whether things are going well. And this is from a guy called Sam Barnes, a talk he did at a very early death called people are weird. And I'm weird. But one of the things he talked about was setting values at the team level. And then measuring as part of the sprint retro, whether the values are being lived up to buy everybody with everybody felt like you were, you were getting enough in terms of being adaptable and evolving and learning and being happy and so on. And then my my final thing I wanted to talk about today is that I think culture add matters a lot more than culture fit. And we'd fall into the the trap of thinking about individuals a lot in tech. And it's super, super easy to do that when you're hiring, right, because you're very focused on one individual at a time, you're trying to pick whether they're going to be the best addition to the team. But we forget that the unit of delivery is the team, the team is what matters most. And so you need to be really, really clear. How is this person going to fit in that in that existing team that you've got? What are they going to bring that nobody else has got? What's what's their superpower, that's going to help add a spike to that team in terms of its capability that it lacks otherwise, because what we're in a really ideal world, what we're trying to do is find a situation where everybody has the opportunity to do what they do best every day, if you can get the best out of everybody. And that adds up to an amazing team, then that's the best situation to be in. And we need to stop levelling people out to equal consistent mediocrity, which is what happens if you tell people only to work on their weaknesses, right? going from bad to Okay, takes a huge amount of effort. Going from okay to brilliant takes just as much effort. And so it's often unless unless something is like a what I would call a controlling weakness. So it stops them being good at the job. Because it's so important. Like you can't have a people manager can't talk to people, right? But for if you're a developer, or you're you have an engineer who can't present in front of big groups, most of the time, that doesn't matter, right, it's not essential for the role. And so instead, let's focus on getting the most out of difference. We need the this sort of shift in perspective and to stop thinking of it as, as if we're interchangeable resource units. And I always tell engineers that if they're called resources by their managers, they should refer to their managers as overhead until they stop it. So feel free to take that into your teams. But we're more like colours or flavours, right? We're better in complement or in concert with each other than we are being thought of as the sort of individual ingredients that never come together. The Avengers are cool because they have they're good at different things. I would watch the hell out of a seven hawks movie. But I think that the Avengers are better because they're good at different things, right? You definitely couldn't send a Hulk in to do what, what Black Widow can do, right? There's there's a definite spycraft element that he's missing. And so what if we think this is people and roles as a matter of casting? How do we assemble a great team with complementary abilities? I think is the much more interesting question. And I am a huge fan of this movie if people haven't seen Ghostbusters 2016 I got very annoyed at the time. This will tell you a lot about my attitude to inclusion. I got annoyed at the time that there were man babies on the internet whining that women couldn't be in Ghostbusters. I saw this movie 11 times in the cinema and I took 100 People with me over the 11 times. So there is a data scientist somewhere in the centre of the Odeon Cinema chain, who still doesn't fucking know why it ran at Leicester Square for 15 weeks. But turns out if you feel the cinema every week, it runs for another week. Week by the time we went the 11th time they were giving us posters and stuff because they were so entertained at US turning up every week to watch it again. But again, they're a team that is cool and amazing because they're very good at very different things not because they're all the same. Also, if you watch this imagining that Chris is is the like the Clark Kent of Thor, it is an even better movie, I highly recommend rewatching it with that in mind because he is he plays the ultimate bimbo and this is wonderful. And so I talked a lot about that kind of inclusion aspect and trying to try to help people be their own version of their best get the get the most out of who they are. But But how do we actually do that? That's probably one of the more challenging things to achieve in a culture right. And I think people are asking themselves these three questions when they're considered Knowing whether to join somewhere whether to stay somewhere whether say yes to a promotion, people do say no to promotion sometimes. And your ability to attract and recruit and retain a really diverse range of people relies on them being able to answer yes to these three questions. The first is, am I expected here is someone like me even expected to be in this team? The number of times I have startups in London come to me and be like, I just don't understand why we can't, can't hire any women. Like your team photo, looks like a frat boy photo. And your only benefits that you list are like foosball and beer Wednesdays are you really that surprised that you're not the most attractive place to work. But But on a more serious note, I've got a disability and I'm on crutches, sometimes I'm in a wheelchair, sometimes some days I can walk. And I've never interviewed anywhere, even in some of the really big corporate jobs that I've had, that I knew in advance whether I'd be able to get into the interview room, if that was a day where I was on crutches or in a wheelchair. Because nobody, nobody imagines that somebody on crutches or in a wheelchair will be applying for those jobs. And one of the best things that we did when I was at Monzo, was we realised we were having an increasing number of neurodiverse people applying to us because we had a great

a great reputation for inclusivity. And one of the scarier things of coming into central London is it's really loud. It's really, really crowded and busy. And so we started, we put a page up that the candidates were linked to that said, it's going to be loud, it's going to be busy. Here's exactly how you navigate through the tube system to where we are, here's exactly what the building looks like. We highly recommend you bring your noise cancelling headphones, if you have them. If you need a rest during the day, you just have to tell us we'll we'll accommodate that any way that we can. Like it's going to be a bit of an overwhelming day, tell us the ways we can help you. And it made a huge difference. Those folks who are in your helpers who are applying to us. The second question is Am I respected here, and I think this is all my differences going to be seen as a feature or a bug. And a lot of this is about training people on unconscious bias, making sure your interviewers are not saying terrible things to people by accident or intentionally. But in general, just helping people be well enough educated that they're being respectful of folks that they're not, you know, something that has happened to literally every woman engineer I know is that they've been asked Are you here for the design role when they come in for interview, and it happens so consistently, there's nothing wrong with being a designer, but it happens so consistently, that they're assumed to be less technical than they are. It's a it's a blight in our industry, right? My my favourite example of this, though, came from a an interviewee behaving inappropriately where my VP of ENTJ, and one of my ENTJ managers were interviewing an architect candidate. And he repeatedly through the interview, when my VPN, who happened to be a woman would ask a question, she would go, Oh, this will be too technical for you holding his hand up like this. And then answer the question to the guy who was the manager who was both more junior and way less technical, which was, of course, they were, they were taking interview notes on their laptop. So they were also on slack to be like live slacking me about this happening. And I was like, should I come pop my head in the door and see if he asked me to make him a cup of tea because I'm a woman. Let's, let's see if that happens, too. But yeah, you know, there are this is where unconscious bias training and those kinds of things can be really helpful. But both in the the initial interview stages, but also just in day to day in your company. And then the third and most important question is can I be myself and be successful here. And I think this is really important, because we know that if you cannot be yourself, if you're having to spend lots of energy masking or code switching or just pretending to be something that you're not, it takes too much energy, it takes away from how good you can be at your job. Stonewall, who are leading LGBT rights organisation in the UK, about 20 years ago did some research that found that people could be 40% more productive if they could be themselves at work two extra days a week. And that these were not people who were not succeeding in their jobs, right. They were despite the fact they had to hide who they were still doing well, and they could have been doing 40% Better if they if they were given an environment in which they could be themselves consistently. And so, with this one, think about what your leadership team looks like and how similar they are to each other. I've worked places where it looked like the leadership team wore matching underwear to go with their matching ties and haircuts right. And it was pretty off For because you looked up and you were like, well, I don't look anything like any of these people, there's no chance I'm ever going to end up on on in that group. And so sometimes, something that can be useful here as well is to just think about how your leadership team presents to the company and to the world. Because even if they may be alike, in some demographic ways, they're probably hopefully anyway, different from each other in other ways. And so sometimes you can get a long way by emphasising the, the differences in paths and similar. So if you, if you have someone in your lead team who didn't go to university, for instance, it's worth making sure people know that it tells everybody else in their organisation who didn't go to uni, that there's a chance for them to get up into that level as well. And similar, I mean, the ideal is obviously to have a truly representative leadership team. But if you lack a truly like demographically diverse leadership teams, and at minimum, you can show that they're different from each other. And they are successful in different ways in order to show that path for people. I'll admit, I've never had someone who is, you know, diverse, and all exactly the same ways that I am to look up to, but I've had versions of that I've had parts of myself that I can see and other people write, because I'm the one that Daily Mail warns you about. I'm I'm a woman working in tech, I'm an immigrant with a job, which I think is worse than if I were living off the state. But I kind of have to check the headlines regularly to be sure. And I'm disabled and neurodiverse. And I'm queer, and my wife is British, I'm over here stealing their women and their jobs, right. But I grew up white in apartheid South Africa. So I'm very, very aware of the like, immense privilege I have, because of the pale colour of my skin. And like I say, I don't think that I have some magical gay fairy dust that makes me make a team better. But teams that are different from each other are absolutely definitely better. They outperform they out innovate, they're more likely to be profitable, they're more likely to, to achieve greater success at a company level. Like there's so much research and evidence that shows that diverse teams win, I'm at a point now where I just don't give a shit about homogenous teams anymore, they're gonna die, that they're gonna go the way of the dinosaurs, they're either gonna be completely extinct, or they'll turn into chickens, and I don't fucking care which it is, right. So so these days, I just care about the teams that are trying to be trying to be more diverse. And so in summary, in terms of making cultural change real when you're when you're growing teams, and a lot of this was from scaling teams over the years, I know a little less of that is happening now. But hopefully, that are still useful lessons. Dry doesn't work for human communication, high performing teams need the conditions for success. So purpose, autonomy, mastery, and inclusion is the framework to remember, inflection points exist. So think, try and find people who are one year, maybe 18 months ahead of you in the journey, and learn from them. Because they're the ones who are going to face your actual problems. It's not the Googles of the world, who are going to face you the actual problems that you have with people observability matters more than testing. So make sure that you figure out how to keep track of whether things are working on. And then culture add matters a lot more than culture fit. If the only thing you do after this talk is make sure that there's a question at the end of your interview process that you ask every interviewer which is how will this person add to our culture, you will see immense value from that. So if that's the only thing you remember, please feel free. But like that's the that's the question to add to your to your interview process. And that's all I had for the presentation so far. So let me stop sharing, and we can head over to the q&a.

James Dong  28:54  
Awesome. Wow, that was that was so amazing, Meri, that was that exceeded all of my expectations. And you know, it's interesting, one of the things you said, there, I've taken so many notes, it will be really hard to summarise this. But one of the really powerful quotes for me that I'm going to walk away from is, when people are looking at this environment asking that question, do I see my differences as a feature or as a bug? That was for me a really profound way of explaining that. I like to open it up for q&a. If folks can feel free to put questions into the chat. But I will also start with some questions that we've collected from folks who RSVP on the meetup. So the first question that I love to dive into from Craig Lewis, which is what are the best things that engineering leaders can do when other leaders disagree on what good culture is, or when other leaders don't really care about culture at all?

Meri Williams  29:55  
So I think when other leaders don't care about culture at all, there's some research and Some papers and stuff that you can grab and try to convince them. But I think that the best way to convince someone is to find something that they do care about, and then make them realise that that is cultural. And key, Houston has a great way of talking about this where she says, engineers think they hate process, but the process they do love they call culture. And so sometimes what, what people will believe in easier than this sort of almost hazy idea of culture is some of the practices that you have as a as a team or as a as a unit, right? And so I think, try and find something that that person does hold dear, and then go, Well, hang on. Nothing says we have to do that, other than the fact that we do it here. So what is that, if not culture? I think when you disagree about culture, that's a much more interesting problem. Because I think then you get into what is fundamentally a question about values. And in an ideal world, your values and your culture match up really well, you're not an Enron example, where they had all these aspirational values on the wall, and then Enron still happened, right? But I think the I think values is the concrete behaviours and beliefs that lead to culture. And the thing about values is, they're a lot easier to include in testing in your interview process, they're a lot easier to look at from a you know, what, in your career framework? Are you encouraging the right behaviours? Or are you showing the right values? So So I would also encourage people to distil down into the values that they think that people have to show in those behaviours that people have to show? Because that's often the best articulate articulation of culture? I'm not is that a good answer? I'm not sure if I've fully answered both parts of the question.

James Dong  31:57  
Great. Got a thumbs up from Greg, thank you. Another question comes in from Andrew Murphy, currently consulting and a company that values autonomy more than anything else, to the extent that they see low mastery as a consequence of autonomy. We are giving people the autonomy to write bad systems, and how would you handle that situation?

Meri Williams  32:19  
That is a that is a fun one. I the way I tend to talk about autonomy is that I want to trust people. And I owe them great context in order to be able in order for them to be trustworthy. And so in that situation, I think what's happening is that people lack the the context that they need on the quality of the system, and the scalability of the system being essential. And so I would have a conversation with with leadership and say, Hey, one of the side effects of us being all in on autonomy seems to be that we're causing poor co code to be delivered. So either that means that people don't know how to deliver good code, or they don't think it's important. And, and I'm making an assumption here that it is important. And the shitty systems are not the ideal outcome. But But I would put it on on leadership to give that context well enough, so because I think if people know what's needed, and they have all the context they need, they can make good decisions, and they can give good actions. And so I think it's probably either about a lack of understanding of what good looks like or about a lack of context that the durability and sustainability of the systems is really important. So and I've actually had that experience, I've gone into some organisations where the technical bar was way too low. And we had to, we had to raise it and having a pretty pretty direct conversation with the whole engineering org to say, hey, I don't believe that you guys want to build stuff this shit as you're building it, I think that there's something missing. And I need you to tell me whether that's time or importance or training, or a combination of the three and then we put in place training programmes and put in place architectural reviews and those kinds of things to not not to police people, but to help them to, to do better. Does that help?

James Dong  34:28  
Yeah, great. Thank you. Thumbs up. So a question from L or Lachlan. What do you do when you have a small team and you need everyone to do everything? So you in that 10 Team situation, but some people are good at some things, but not at others?

Meri Williams  34:42  
I think the ideal is you map it out and say you know what, by our powers combined, we are Captain Planet, right? We've got you know, let's let's organise ourselves so that people get to do what they're great at at any given point and then, but let's figure out how we have a second In command for every, every one of these things, so you don't want somebody who can only do one thing and they're sat idle the rest of the time, you say to them, you're brilliant at that, we'll use you for that. But I both need you to train somebody else to be good at that. And I need you to learn to be good at something else. So you've got to be somebody else's second. And you've got to create a second of your own. And I think then you get the the learning journey, but you also get the redundancy that you need, if somebody goes out sick or on holiday, which, you know, hopefully they should be allowed to do, then then you've got that. So I don't think that we're that doubling down on people's strengths is about ignoring things that you know other things that they could get good at. But it's it's preferring to say, let's take you from good to brilliant or okay to good, over let's take you from terrible to okay, if we have a choice, and the thing you're terrible at doesn't matter so much because somebody else is already good at it.

James Dong  35:58  
Great question from Briana, once the major problems have been addressed, how do you like to use retros and make the most of that process once fewer things are on fire?

Meri Williams  36:11  
I think so. So two things I like to do, it's a great opportunity for people to give each other positive feedback. And so I think switching up a retro to be celebrating it as well as as well as investigative is useful. And then the the other thing that that can come out of it is that you encourage the team to be less internally focused and to think more about the broader picture. So is there something in how a lot of times I'll see teams when they do a retro? They're very focused on what's within their sphere of control, right? So they're like, What can we change within the team to make things better, but you can get to the point where everything's working great within the team. But I bet you there's still stuff that's wrong with with collaboration with other teams with the fact that other teams aren't working as well as they are working and that kind of stuff. And so encouraging them to think about what else can they not control, but maybe influence can be can be useful as well. But the main thing is to make it very celebratory of the of the progress that's being made week by week, because like we saw in those questions, right? It's every seven days, I've gotten good, good feedback for for great work. That's a that's a pretty challenging target. And not one a manager alone can meet.

James Dong  37:27  
Yeah, that's a great insight. Another question from Janet, how do you create mentally and emotionally safe spaces in work culture, particularly when leadership might have the right values but never been exposed to these new ideas or methods?

Meri Williams  37:41  
So I think that I've done with leadership groups is to take this, am I expected respected, and can I be myself and be successful and give each person in the group a different hat to wear, and they get told in advance what that hats going to be. So if they've got to represent Muslim teetotallers for the day, or you know, for the workshop, or you know, somebody in a wheelchair, or somebody with ADHD, or person of colour, or you know, whatever the, whatever the appropriate mix is, you ask them in advance to go and really think about that, hopefully find a friend that they can talk to you to understand what kind of challenges that person has in the workplace, that kind of thing. And then you bring them into the room and ask them those, those three questions and say, what's in the way for you with the persona that you're in charge of today? What's in the way of, of you feeling really expected? There's no pictures of anybody who looks like this on our website, there's no indication that people like, you know, we don't have a prayer room. So if you're an actively practising Muslim, maybe you don't feel very welcomed in in advance when when you see our you know, story about our office and who we are and what we are or whatever, right? Are you respected? Do we have any instances where somebody would would feel like their difference was was a bug rather than a feature? If you don't drink then huge number of cultures are that that feels like a bug rather than a than a feature, right? And then can I be myself and be successful? genuinely trying to figure out for all those different kinds of people? How would what signals does the organisation give them not? What signals is the organisation tell them that they're not going to succeed? But what signals do they get that they will succeed? Because realistically, if you're in any kind of underrepresented group, the world is giving you a shit tonne of signals that you're not going to succeed all the damn time. And so even a neutral environment organizationally, is not neutral. It's negative, because the context that we're in is so negative for the for all of us who are in underrepresented groups, that we can assume that no information is good news. No, inflammation is just as bad as the rest of the world. And so I think that's that's a specific exercise that I've done with a lot of leaders that that seems to help.

James Dong  39:59  
Great, amazing Answer. I'm gonna switch it back to online to ask these questions from Stella, could you share some anecdotes about balancing culture building and driving results? Especially when sometimes they might not align? And how do you make that trade off?

Meri Williams  40:14  
So I, I think in an ideal world, but it's an ideal world, your culture is not at odds with delivery. And, and with getting shit done. Because like I said earlier, people love to be great at what they do. They don't want to wake up in the morning and being mediocre, they want to be great. And so ideally, figure out the ways that your culture is holistic, and includes delivery and includes getting shit done and includes making things happen for the customer or for your colleagues or whoever, whoever is appropriate in your particular happenstance. But I get what you mean. And there's, there's times when you know, maybe you have an aim to have a really balanced culture and good work life balance, but then there's a deadline, and you're asking everybody to work evenings and weekends, right. And so I really go back to that, that quote about the standard you walk past as a standard you accept, I think the things that we do, show what we care about. And so if you say that you care about work life balance, or people's personal, you know, having having a right to a personal life, but then you repeatedly only celebrate people who work late and work weekends, then that is what your actual culture is. And so I think that requires a change in what people reward. And it and some of that's just a simple change in, in language by leadership, right? That's saying, Thank you for doing this. We know it shouldn't have happened, we'll do our best to make sure it doesn't happen again, rather than thank you to the heroes who jumped in and did everything. And that's, that's all but you say? This is a really big question, though. So I think I'm only scratching the surface of answering it. But hopefully, that's a little bit helpful.

James Dong  42:09  
Great, thank you. Question from Zeke, back in the chat. How do you keep a healthy ritual culture from becoming a stale and mechanical ceremony?

Meri Williams  42:19  
I think limiting the number of action items people can take, because I think what happens in a lot of retros is that everybody goes, this is everything that's wrong. And what could we do about all of it, and then, like, a week later, or two weeks later, not everything is changed. And so it feels like nothing has happened, right? And so you have to go, what's the one thing we're going to pick to change. And actually, the best way to convince engineers at this is to go, we've got to be scientific about it. Kids, we we if we change multiple things, we don't know what worked. So let's pick one thing to change, let's have a proper hypothesis. And let's change it and see what improves. And I think that's the key. Because if you if you try to take too many actions out of every retro, it's just too easy for it to become horribly repetitive and really demotivating for people that they feel like they're saying the same things over and over again, but it's never getting better. And so sometimes it's you know, you run a retro and it's a everything that's wrong in the world, kind of retro and that's an that's appropriate. But then I think you guide future retros to be a little more focused around the specifics. I also think that changing up the the actual retro format that you're using can be really helpful. So I've the the Jessie links talk, that's in my slides. She even looked at some of that she used some methods like drawing the sprint, which I am nowhere near visual enough to do that successfully, but apparently worked really well for one of our teams. And, you know, there's lots of different styles and formats that you can use. And so changing up what style and format you're using, uncovers different things as well.

James Dong  44:01  
Awesome. And this is a question from Tang in the chat. And then also online, how would you make a case to an organisation to start investing in and hiring more juniors? And then specifically on the hiring question, When might you determine the best time to hire a person to be

Meri Williams  44:20  
so I think in terms of getting more junior folks in the, the main arguments that I tend to use, they'll tend to stay longer, they develop rate, expertise in the company and like lots of great contexts. And if you can give them a good start in their career, then they'll typically give you more like three or five years, where average tenure in in more senior roles tends to be 18 to 24 months. In Europe anyway, your mileage may vary in other parts of the world, right. And so and then the the other argument I use is that we can't have people who are senior unless they're coaching people who are more junior than themselves. So if you have a bunch of alleged seniors who have nobody to coach, then they're not seniors, they're just mid levels who have a lot of responsibility. Because they're not showing the full range of behaviours and expectations that you expect of senior engineers anyway.

James Dong  45:19  
Right. Got it. And a question from Stephanie, which is, how do you feel about the current trend of seeing your work colleagues as family? And what are some of the impact of that on the modern workplace and on leaders?

Meri Williams  45:32  
I think it's horrible and toxic. So because you can't lay off your family, I you know, I'm, I was a queer kid growing up in South Africa, I would have loved to lay off my family. I was not allowed to. I think it's really toxic. I think that having your work environment be somewhere where you care about people, it's fine. And somewhere where you can have friends is fine. But I don't think that your work environment should be your found family. And I think it's a pretty i i can understand where some like founders and stuff are coming from where they, they hope for that rain for that, like, I've actually worked for someone who said, like, I started a company because my parents got divorced when I was 18. And it really messed me up. And I wanted to do, I wanted to create more stability around myself. And then he he is the CEO that I've worked with who takes the most personally, when somebody quits, have anybody I've ever met, everybody in New York, like one of the factory workers would quit, and he'd be like, dead to me now kind of thing. It was just, it was it was really weird. And so I think it can be really, really toxic. I don't think it's a good, good way of framing things. And I think you can get a lot of what people actually like about that, you know, about treating people as humans and bringing your whole self to work and being able to be yourself amongst your colleagues, I think you can get that without having to say that you're like a family or that you are a family. So it's, yeah, I'm very against it, I'm afraid.

James Dong  47:10  
I'm gonna ask one last question. And it is a combination of two questions that we got, which is how do you foster a culture of feedback? And specifically for engineers? How do you help engineers be more open and resilient and less fragile to constructive feedback?

Meri Williams  47:28  
So there's a, there's a really good book called thanks for the feedback. There's a, they did a talk at Google that I would also recommend watching. And actually having the whole organisation trained on how to give feedback is very useful. So telling people, here's how you give feedback, focus on the action and the impact that it has, don't tell somebody, they're a bad person, or, you know, don't don't give feedback that makes it sound like a characteristic of themself, but give feedback on the actions that they take the behaviours that they show, I think can be can be helpful. Don't use the shit sandwich, don't tell people a good thing that a bad thing than a good thing. Because everybody well, one of two things happens either you get people bracing for bad every time you say something good, which is not a good outcome. Or people are like two out of three ain't bad all door the bad thing, depending on how optimistic or pessimistic they are, right. And so I don't I don't think the the shit sandwich is a good idea either. And, a lot of times when, when an engineer is super defensive, it's something that they're already aware of. So the other thing I would suggest in in one to ones is create space for one to ones to be more like a personal retro. So ask them what do you think's going well, what do you think's not going so well, and actually, if they bring up the bad, the bad behaviour first, it's a lot easier for them to be less defensive about it. And so helping people get to the point where they trust you enough to be like, Well, I'm, I'm worried that I'm not doing great at developing up our younger colleagues and my pull requests, reviews or whatever, right? If they bring it up, it's a lot easier. And it can help you know, whether they've got that self understanding, and they're just not sure how to work on it, or whether they're oblivious as well. If you can create that space in which they can be a little more open about about what their what they need to be better at. The other thing I think that is useful is to try and get the team to share with each other what they're trying to get better at, because I think that normalises needing to get better at things. And then teams can support each other as well. They can you know if I know that you're working on giving more constructive feedback and pull in PR reviews. Then I can like note that you did great at that today and drop you a quick slack and say hey, that was a much better than usual one thank you for investing the time in it and then and then teammates can help each other with positive reinforcement as well.

James Dong  50:11  
Great, amazing again, thank you so much. It's really clear to me hopefully it's clear to everyone here that Meri really walks the talk on doing culture and doing diversity equity inclusion, actually fun fact. And Meri, I read this in an interview or or another talk, so I think I can share it. But I know that for you is really important to keep this chief tech Chief Technology Officer title, rather than product because you want it to be an example of a really senior technical person. So that's really amazing. And I'm going to stop the recording but then we have a couple of last things to just wrap up.

Transcribed by

James Dong
James Dong
Support your developers with ethical team analytics.

Start your free trial

Join our beta
Support your developers with ethical team analytics.