The interview lasts around 45 minutes, followed by a 15-minute Q&A. If you don’t have time to watch the full video, you can use the timestamps below to identify the parts you’re most interested in. Alternatively, you can search the transcript of the conversation for particular keywords using Control+F (or Command+F on a Mac).

Resources mentioned in the interview

Timestamps

03:38
• cracking the viral coefficient at Skype
• integrating with Windows, Mac, and others
• design systems/GUI kits
• Skype was bought and sold three times, ending up at Microsoft

11:27
• the value of a design system: build once, use often
• Skyscanner’s first design system was called “living styles”
• control of experiments and partner brands’ themes

14:33
• innovation/differentiation in the travel sites space
• business goals geared towards optimization and conversion
• frequent experimentation, multivariate tests

17:48
• arguing against pressure messaging
• culture at Skyscanner: do the right thing for the customer
• doing the right thing for the customer builds trust in the longterm

20:55
• building design teams that believe in the design principles
• evangelizing user testing
• emotional components that show you understand the customer
• prototypes, concept cards, North Star metric

28:05
• defining what not to do
• design leadership vs. business goals
• leading from the front as a role model (do as I do), delegating
• each squad needs an understanding of who leads what

33:07
• inspiring from the front, leading from the back
• servant leadership
• the three As framework: accountability, autonomy, alignment

36:21
• fixed design processes vs. creative freedom
• processes provide predictability but not a proxy for success
• double diamond process, processes define types of feedback
• directive vs. prescriptive feedback

42:06
• managing multicultural design teams
• challenge of different cultures, languages, design nuances
• the accountability ladder

45:31
• incremental design changes vs. reimagining the entire experience
• building the thing that kills you before someone else does
• weighing operational risk, operational parameters

50:46 (Q&A)
• Can you build a culture from the bottom up by influencing directors to take it seriously?
• Do you have a specific process to feed customer insight back to the business?
• How does Skyscanner use qualitative and quantitative research data?
• How did you go about understanding how users make choices about flights?
• How did what you learnt from them affect your team’s design choices?
• Favorite resources?
• Why is Steve called Buzz?
• Any final advice for aspiring design leaders?

Keep up with Steve ‘Buzz’ Pearce on Twitter

Transcript

Sandro: So, let’s get to it. Let’s quickly start with who we all are, and why this whole webinar series exists, and who is behind it. First of all, this is the first episode of season two. Season one, if you’ve already been part of it, was called Fireside Chat. Now we re-branded it to Let’s Talk UX. As I said, this is the very first episode, so welcome, writing history together with us.

Sandro: Basically, we’re still the same people behind it. First of all, me I’m Sandro. I’m working at TestingTime here in Zurich. We are a study participants recruiting agency. We have a pool of 350,000 people. If you need study participants in the UK, Germany, Switzerland, France, Netherlands, we can deliver them to you in 48 hours. Our partner for a couple of years now already is UX Live, and the partner company behind UX Live, Tech Circus.

Sandro: Actually, UX Live is a big conference that’s happening. It’s like a three-day conference that’s happening in London in November, so this year it’s going to be from November 11th until November 13th. Actually, Steve is going to have a talk there, and so this is kind of precursor to wet your tongue a little bit for what an interesting person he is and all the stuff he has done. It’s going to take place at the ExCeL center in London.

Sandro: Actually, Mike from Tech Circus will share the link with you in the chat in a bit. Basically, what is it UX Live? There’s going to be many speakers. There’s going to be workshops. It’s like leadership stage, a participation stage and a panel stage. It’s mainly a conference for UX researchers and UX designers. But you can check out the link, and there’s also a small offer at the end where you can get the tickets a little bit cheaper, so I will share that with you at the end, so bear with me there.

Sandro: Okay, that’s for that, and then now to you, Steve. I have to almost read your accolades here, so that I don’t forget any. Steve or ‘Buzz’ how they call him, we’ll learn later why, has built some of the most loved products and brands on this planet I would say, so there’s Lego, there’s Alexander McQueen, the eccentric fashion label, Jamie Oliver, Skype, Microsoft. He worked at Microsoft after Skype has been bought. He’s also on the board of the Design Council of the UK, and then he’s board of directors of the Design Business Association.

Sandro: Actually, one funny thing I saw is that Steve has designed some of my favorite … I hope you see my screen here, some of my favorite emoticons. I always really like this one, and I always really like this one. Steve, basically, when you joined Skype, which apps have already existed because I definitely remember the desktop apps, but then when I browse through what you’ve done on this website, I saw there is apps for Skype, for TVs. There was one integration with Facebook. Then there’s, I think, for Xbox as well, and then you have for all the different mobile phone versions out there.

Sandro: My first question would be for you, how did you go about that? I mean, we talk a lot about design systems today, and atomic designs and all this kind of stuff. You started at Skype, and then how did you think about that? Welcome first, so welcome, Steve. The mic is yours.

Steve Pearce: Thank you very much, Sandro. It’s a pleasure to be with you all. I hope you can all hear me okay. Thanks for the questions. I hope this will be interesting for folks. I’ll be very honest and up front about my failings as well and the things I’ve learned. Skype was a bit of a rocket ship. You probably remember back in the day it had probably been the most fastest-growing app of all-time around sort of 2010.

Steve Pearce: We had very early on cracked what is called the viral coefficient. The viral coefficient is basically how many people does one user bring into the system, and you’re looking to get that as close to one, so you can never Skype alone, so of course, our growth was double digits. We didn’t actually have to do that much because we cracked that early on. Of course, when you’re looking at where people are using the app and their demands, we were very clear about listening to what apps and services people wanted.

Steve Pearce: At that time, the iPhone had only just come out, but it had no front-facing camera, the hardware being blunt wasn’t quite up to the standards that we needed for Skype to work in a performant way. Mobiles up until that point had been integrated largely through Skype Credit, which was you could actually be integrated onto the most basic of phones. They were hugely popular because you didn’t need a contract or anything at those time.

Steve Pearce: What we rapidly realized as soon as we started going just beyond Windows into Mac and then onto other platforms that wanted these integrations is, like you said, you needed a design system. Design systems hadn’t matured at that time, and they weren’t necessarily called design systems. They were just called GUI kits. We had numerous repositories for the GUI kits for the platform that we were integrating, for example, Asus Videophones, the sort of things you’re now seeing from Google and Facebook where it’s just a pure VC machine with a calendar and an email.

Steve Pearce: We had done all of those very early on, and they were hugely popular. The GUI kits had inherited certain properties of the platform, particularly when you went into TVs and onto very low bit basic phones. Some were even black and white because they were just doing audio calls. What we had to establish was not necessarily the GUI kit, but actually it was the app framework and the various states that were very much needed to run a basic version of Skype, so you needed a contact list, you needed a chat state, and you needed a live state, and then you needed all the bits that joined all that together, which was adding someone or the arrow mechanisms, and then you need all the management capabilities.

Steve Pearce: That was really a large portion of the entire team’s work was devising the app architecture with engineers. Once we had cracked that, it was fairly easy to say what is the minimum required to run Skype on the most basic of devices, and then what is the maximum supreme experience that we had want. At that time, that was the desktop apps. They had the most horsepower, GPU, CPU, et cetera, in order to run and do decoding on the device because it was a peer-to-peer network, so really, really impressive stuff. I can’t speak highly enough of the amazing engineering talent at Skype at the time. They were real trailblazers.

Steve Pearce: But designs at that time was devising the GUI kits and the app architecture. As we matured, and you know Skype’s history just a little bit, I suppose we were bought and sold three times; eBay, then to private equity, Silver Lake, and then finally to Microsoft. It wasn’t really until we joined Microsoft that we started our design system in earnest when we started devising Skype for Web, which is plugin-less Skype equipped. We devised a system called [Mata 00:09:32], which you’ve probably never heard of. It never got made public, but it was very, very impressive if I do say so myself.

Steve Pearce: I did not build it. The team built it, some wonderful minds that are still at Microsoft. One of the first things that we did was to create the motion language and get that into production ready code. I could wax lyrical about this all day, but I can give you some names: [Joe Philips 00:10:02], [Andres Collart 00:10:02], [Machek 00:10:02], [Bobby Raitt 00:10:02], [Martin Crochet 00:10:02]. So many of these amazing brains actually devised plugins for the likes of After Effects, for Illustrator, for Sketch that spout out production-ready code for engineers to implement.

Steve Pearce: You’ve probably heard of Lottie, which is basically the public version of what we created, probably slightly better under the hood admittedly. That was really the beginnings of a design system where engineers and designers were really collaborating to make the UI as consistent as possible across the platforms. Of course, we had to do that integrating with Microsoft at the same time, which was going through a massive rebrand in their Microsoft design language. We were part of that team to help devise that, so many, many moving parts.

Steve Pearce: My second office was seat 60K on BA87 to Seattle. I was there all the time. The complexity was exponential in terms of what the design system had to support across basically almost every platform on the planet.

Sandro: At that time, people just began to really fathom what designs systems meant, and nowadays even smaller companies are also thinking about that. How has that changed now that you are at Skyscanner? How has that changed? I mean, when you started at Skyscanner, was that all there? How much did you need to influence what you’ve learned at previous positions and in that regard? What was the situation there?

Steve Pearce: Well, one of the things that we learned at Skype and Microsoft was the value of a design system to help give some sense of guidelines across all these disparate platforms. When I joined Skyscanner, there were two people working on the first iteration of that, and it was called living styles. As I recall, it was CSS and mixins to basically try and bring some sensibility to all the different button styles. It started very much in its infancy. Of course, the work of Brad Frost in atomic design was a real helpful metaphor for the business to understand and for engineering to understand what we were trying to accomplish.

Steve Pearce: And then when I came in it was really evangelizing that at exec level as what were the benefits, and the main benefit there is speed to delivery and cost saving, because you build once, and we use often. As it matures and as it develops, we realized very quickly that we needed a slightly different mechanism to control the experiments that we were running and to also control the themes. The themes were being derived from the partner brands like Virgin Atlantic, or Delta, or British Airways where we integrated their brand into our brand, so you needed to do run time theming, which was quite a breakthrough [crosstalk 00:13:50].

Sandro: Sorry, how do I understand that?

Steve Pearce: When a person lands on Skyscanner, and you want to buy the tickets from British Airways, we can theme the page according to British Airways branding rather than ours. At the very high-level you have theming switches, and those people that know design systems, that’s quite an important thing to be able to do. Yes, it’s been very successful. You can all check it out on backpack.github.io. That’s our public-facing, and you can pick up and use that design system if you want.

Sandro: Exciting. I mean, now that you’ve switched to the travel space, what would interest me is how do you think about innovation in the travel space? When I look at a lot of the travel sites, they all look kind of similar. How do you differentiate? How do you think about that topic?

Steve Pearce: Well, that’s a good topic. I think one thing that we struggle with as a design community is the business goals are very heavily geared towards optimization and conversion of customers. One of the issues that I think, and this is my opinion, not necessarily the opinion of Skyscanner, so let’s be clear about that, is you have a frequency issue. Most people come to travel sites, regular folks, once or twice a year.

Steve Pearce: There is a huge temptation to therefore bring out your entire arsenal of weaponry to close that sale, which leads to what I call [perma 00:15:49] anxiety, “Am I really getting a good deal? Can I really trust this information? There’s all these peripheral messaging, pressure messaging that comes in as well like, “Hurry, only two seats left, two rooms left.” I think that taints the entire industry quite a lot, which doesn’t lead to long-term trust.

Steve Pearce: The other part is because there is that desire for the business to increase conversion and so on, you end up going into experimentation where you basically do multivariate tests on almost everything in the funnel. Some things work, some things don’t. But when people are going to an average of … I think it is about an average of 20 different travel sites before they make a booking. No wonder the patterns that each of them compound will have an effect on the next one and the next one, so you end up with this kind of global homogenization due to experimentation where they all end up looking the same because the pattern and muscle memory associated with each one it’s just compounded and compounded.

Steve Pearce: It takes a very brave company to break out of that mold, but once one company has done that and really started pushing, then other start to follow suit, but it takes a very brave business to do that because how long will a business keep going when the numbers aren’t necessarily increasing? Of course, the roll back to a previous version is quite pronounced.

Sandro: Because you still have to sell something … Frankly, I told you before this talk when we had a quick chat I was quite surprised when I actually went to Skyscanner and actually looked, “Okay, there’s only two rooms left, and you have only 10 minutes and stuff like that.” How did you go to your bosses and have this conversation where it’s like this conversion rate, optimization business part versus the long-term effects of your user experience? How does a conversation like that go on, and how would you recommend other businesses to have this conversation inside their company?

Steve Pearce: It’s a very good question. One of the reasons why I joined Skyscanner is their culture. Their culture is obviously made of a set of principles and values. One of the key ones that was very apparent was that the business always wanted to do the right thing for the customer. Now, that’s easier said than done when you have to hit business targets, but one of the key things before I joined as a proof of this, and this is the ultimate proof whether you really do put the customer first, is how much are you prepared to do the right thing for the customer and hurt the bottom line of your business?

Steve Pearce: I’ll give you one example. When doing a search on Skyscanner web, and this is four years ago, you were able to do comparisons with other sites, which meant London, New York, here’s the dates, search, and there were just some check-boxes, which was, “Do you want to compare with lastminute.com and with Kayak and so on?” We would charge Kayak and so on an advertising fee for popping up their website under ours as soon as you hit search, and users, guess what, hated it. Absolutely hated it. So many complaints, “Why have you opened up all these browser windows? It’s spamming me, et cetera.”

Steve Pearce: But it was generating tens of millions a year, but we said, “No, this is not good for the customer. It will not earn long-term trust.” That philosophy has been maintained. Therefore, internally and externally we always say we are the traveler first travel company. We will always put the traveler’s needs in front, so when it comes to things like pressure messaging, hurry, only two rooms left and that kind of thing, we won’t do it because it doesn’t help. It might get you short-term better conversion, but then ultimately every thing starts to have to shout.

Sandro: Can you talk more about the principles that you apply inside Skyscanner? I think more importantly then also because everybody can have principles or write them down somewhere, but how do you build teams around those principles, and how do you hire new folks that actually are in line with those principles, and how do you make sure that those are consistently applied even when you are not there let’s say?

Steve Pearce: Well, one of the best ways to affect behavior is to say that you’re going to be measured on the behavior. Whenever we do performance reviews, it’s how much has this person or this team demonstrated they have put the traveler first. That’s one thing. The second thing is you have to evangelize all the user testing that goes on, so you get people as close to the customer as possible.

Steve Pearce: Now, I know that I’m not saying anything particularly revolutionary or new, but when you’ve got a company that does have a very clear understanding of itself and its principles, there is no point in having principles and values if you’re just going to break them. You have to uphold them, and you have to constantly repeat them, and you have to constantly demonstrate why something is good or bad for the traveler. That, in turn, brings out an internal mantra that then becomes part of the DNA of the company.

Steve Pearce: For example, internally, our DNA is traveler first, partner second, Skyscanner third in that order. There’s always a balancing act. We’re still running a business, but we will always default to putting the traveler first because that’s how you ultimately build long-term trust. One of the aspects I will just say, when I actually started to have this conversation about how do we codify it with the leadership in the business, is what is our Achilles’ heel as a business? What are we actually trying to fix as a business model for the traveler?

Steve Pearce: I’m sure you’re fairly familiar with Skyscanner. The way I characterized it was like a wonderful calculator. A calculator is input, output, you get the result, and it goes back in the drawer. It’s not very emotional. It’s very functional. That’s all good. That gets you to a certain level, but now put that in the context of travel, which is what’s called an asset-heavy business, which is largely around your memories, and emotions and so on. It’s not just functional. It’s incredibly-

Sandro: It’s a bit like fashion or food.

Steve Pearce: Yeah, it makes you up, and it makes you the character that you are when you experience other cultures and so on. Having that blend of functional and emotional is really important, and, of course, that’s where designers really love to bring those things together.

Steve Pearce: I will remember a company meeting where I gave a presentation to the whole company. It was about making ourselves more than just a calculator. We want a relationship with our customers. You don’t have a relationship with a calculator, so what are the emotional components that we will start to layer in to show that we actually really do understand travel and we understand you the customer?

Steve Pearce: That was a fairly straightforward and easy conversation if I’m being blunt. It just required the design team to actually show what that might look like and how that might work via prototypes, and you might call it concept cars where you can paint the North Star of, “Hey, this is where we think the experience can go,” so then you start to break it down, and then you end up taking the business on that journey as we go, but we’re more than just about flights.

Steve Pearce: It’s actually going to be about trips because we’re doing flights, and hotels, and car hire and et cetera. Therefore, the experience must shift from being flight-centric to trip-centric. Okay, great, because then we start to get a flight booked and then a hotel booked via a touch. If you organize your model around that, you actually then can accelerate the flywheel of the business. That’s just one example of how one type of conversation went. Other conversations were far more emotional. I can give you one more example.

Sandro: Please.

Steve Pearce: This one brings a few wry smiles to folks that heard me talk about this, but if I were to do it with the audience here, it would be hands up. Who has a partner? I’ll do it with you, Sandro. You have a partner?

Sandro: No, not of now.

Steve Pearce: Okay. Well, then it falls a little bit flat. If you were to have a partner, my next question would be, do you love them?

Sandro: I surely would.

Steve Pearce: You would. And then I would say, “Prove it. How would you prove that you love your partner? Show me the algorithm. Give me the number.” You say, “I don’t know. That’s not how it works.” Of course, that’s not how it works. You would say, “Look at my actions. Look at my behavior. Look at my attitude towards my partner. Look at the things I do for them, how much time I want to spend with them, how much time I want to understand what their hopes, fears, dreams are.”

Steve Pearce: Okay, that being said, let’s flip that round to the business. We put the traveler first, right? We love the traveler. We love our customer. Prove it. What would you do? What little thing would you do? In a relationship it’s not always the big gestures. It’s the little things you do every single day. Think about that every time you’re designing a system. How will it make the traveler feel? How will it demonstrate that you love them and put them first in everything that you do?

Steve Pearce: That seem to hit a note with the majority of the company. Often it’s just little stories, and it’s a way to sell those seats. What you’re doing is you’re stepping up a whole grand of universal principle of what the business purpose is all about, which is to put the traveler first, and then you’re setting out your vision, and then you’re setting out, “Well, if that’s our vision, then how will we go about doing these things?” That’s your values, the things that you hold dear, the things that you won’t ever do, because often I find that it’s the things that you say you won’t do are the more important things than those things you will do because that gives you the guard rails. And then you start develop a design language that reflects that.

Sandro: Do you have like a not-to-do list? Do you have like a list that says like, “These are the things we never do,” or is that implicitly given?

Steve Pearce: Well, I’ve given one already, you know, things like putting pressure messaging in our apps. You start to develop these things as you go, and they become a part of the DNA. Never gossip about someone behind their backs. That’s another value.

Sandro: I mean, you as a design leader you still have to make business decisions. How do you think about your role, let’s say, as a manager versus as a design leader? You know what I mean? The business aspects of it versus actually shaping these cultural things that you’re now talking about? I guess many people in the audience have a similar problem. On the one side, you have these principles, and then on the other side you have these business needs. How do you solve that tension?

Steve Pearce: Well, there’s always going to be that tension, so you have to acknowledge that. That’s where you have to start making hard trade-offs, and the business will always have to do that because it’s running a business. As a leader, you’re ultimately putting your customer first, but you’re still running a business. You’re going to always have those tensions. There are very rigorous and long debates about how to make those trade-offs.

Steve Pearce: When it comes to having your sort of which hat you wear at any given time, the management side is really about the performance of the people and setting up the environment for them to flourish. Leadership, in my opinion, is actually just simply this. It is to lead. It is to lead from the front. Effectively, you are the role model, so it’s do as I do, not just do as I say. You’re setting up the principles for how management will operate.

Steve Pearce: When we looked at this, and this is something that we have devised as a leadership team at Skyscanner, is sometimes people are actually quite reticent to say what they need and what they’re going to be held accountable for. We have a very simple framework, which is in a given team, and we run tribes and squads. In a squad you’ll have a group of people, different disciplines and so on, and it’s very important that squad has an understanding of who leads what.

Steve Pearce: We come together. Before the meeting, you’re given … I think we use Trello, and there’s a column which says I did, and I support, and I can’t let go of. And each person does this-

Sandro: That’s a good one.

Steve Pearce: … blindly. Then you come together. You then have this discussion, which is, “Oh, I thought I was leading that. You say you’re leading that. Okay. If you really want to lead that, you lead it, and I’ll support you.” And you go round, and you hash that out. It’s very important because then you go, “Well, if I’m leading it, then I’m accountable for it, and you can hold me to account. If I say I’m going to do it, then I do it. And if I don’t, then on me it is.”

Steve Pearce: And then there’s always these things that come up, which is what I feel I can’t let go of. Sometimes there’s these emotionally things that [crosstalk 00:32:18], “If I let go of this, the business will fall apart.” You have that discussion, and then it’s often, “Oh, really? If you did stop doing that, what would happen?” You realize actually there’s a lot of things you can just let go of, and it’s okay. There’s other things that actually if you did let go, the business would fall apart, so you have to have someone on them.

Steve Pearce: Often those things are around processes and governance because that’s really important that you have an operational model that keeps to monitor the health and well-being of the team at large, so a very simple framework. We try and keep these as simple as we possibly can, so that we can grow as fast as we possibly can, but that’s just one example of how we view the different dimensions of leadership.

Sandro: I actually really liked one of the quotes that I read from you, which was like, “Our mantra.” Maybe you can say like, “Inspiring from the front and leading from the back.” I don’t know if you heard that from somewhere or if that’s yours, but I really like that.

Steve Pearce: Well, I think sometimes you do actually have to just lead from the front. There is a general principle, which is people are much more easily drawn than they are driven. That quote is probably few years ago now. I might modify it since then. I think my natural disposition as a leader is what’s been characterized as a shepherd.

Steve Pearce: If you were to look at a shepherd, the shepherd will lead the sheep out to green pasture, and whilst the sheep are grazing and so on, the shepherd retires and allows the sheep to autonomously do their job, and then leads them onto the next field and keeps watch over them. I suppose that’s my metaphor in that respect. It’s not cracking the whip from behind like driving people. No way. It’s not that at all. It’s in servant leadership style.

Sandro: I guess it’s like knowing when to be in the front and when to retreat into the back a little bit as well.

Steve Pearce: Yeah, and I think that’s important when you start defining what I have called the three A’s. Now, this is not just me that’s devised this. There are many brilliant minds here, and we’ve over the years assembled this framework. The three A’s are this: accountability, autonomy, and alignment. You need all three, but which one is the higher priority over another? We have asserted that accountability is the primary thing that you want in people, that they will do the things that they will say they’ll do, and they will own it.

Sandro: They take ownership. Exactly.

Steve Pearce: Exactly. That’s the primary one. Fine. Then the next two is autonomy and alignment. Now, I will always favor giving someone more autonomy. There is never such a thing as 100% autonomy. No one has that. You’re always accountable to something or someone, and you’re always trying to align to the business goals and objectives. The default is autonomy, knowing that they will self-align because they are accountable.

Steve Pearce: Now, that sounds a little bit kind of circular reasoning, but you have accountability, and I’ll talk more about this in my talk at UX Live, but it’s been very helpful for people to understand what is expected of them. I won’t spoil everything, but that’s just the basic outline of the framework.

Sandro: Also, you mention before putting processes in place also in the design world. How do you make sure that while you have processes in place the creative freedom of the team is not undermined? You kind of leave some room for creativity.

Steve Pearce: Yeah, absolutely. Processes are good. Where I don’t like processes is where you use process as a proxy for success. In other words, while we follow the process, why hasn’t it been successful? You can’t do that kind of … That’s the wrong reasoning. Processes are there to help people and keep the business in line in an aligned fashion and make it predictable for the business to understand when things will be done. That’s all that a business ever wants is some level of predictability. That’s why Scrum, and Kanban, and Agile have been so popular it’s because you can start to get people’s cadence, and velocity and so on, so it gives some sense of predictability to the business.

Steve Pearce: But what we have done is whether you like it or not, and there’s no particular creative framework that’s perfect, is we looked at all the different frameworks, and process, patterns and so on, and we just very simply and lightweight adopted the Double Diamond as the most basic way of explaining to the business where we are in our process. That’s been actually very helpful. Within the first diamond, you can then exercise your creative freedom, which-

Sandro: Like you open up.

Steve Pearce: You can open it up. All we’re saying is, “Look, we’re exploring, we’re diverging, and in two weeks we’re going to start to converge, and we’ll come to this point where we need to come together, agree, disagree, and commit to the solution that we’re then going to go ship.” We try and do that as quickly as possible. That’s it really. It’s not trying to be too clever about it all. It’s just from a reporting perspective where you are in the process.

Steve Pearce: One thing though to know, and I’m sure design folks will be interested in this, but you have product design reviews, you have critics, you have peer designing, you have work in progress sessions, you have all hands. You have all these ceremonies and rituals. One thing that is important I think is that you need to explain where you are in the process in order to elicit what type of feedback you’re really wanting.

Steve Pearce: For example, in the first stages you’re probably about, I don’t know, 20% on the journey in which case the feedback is going to be quite high-level. It’s not going to be prescriptive. It’s going to be directive, “Yeah, keep going in this direction just a little bit here.” It actually goes through that, and it gets closer and closer. Things start to get way more precise and specific and often prescriptive, “Move that here. Do that there. Do that here.” You got to be very clear on what type of feedback you’re asking, and therefore what type of feedback you’re going to be given.

Sandro: What you’re saying is whenever you want feedback that you’re clear about where you are in the process.

Steve Pearce: Absolutely.

Sandro: So, you get the feedback that is actually appropriate.

Steve Pearce: Absolutely. Very, very important. Of course, it’s up to the person receiving it whether they want to receive it or not. I think as a leader the thing that I have learned is when to be directive and when to be prescriptive. Often the prescriptive comes when things aren’t necessarily being aligned. Therefore, you can say to the team, “Okay, what I need to do here is align you to this, and this is the reason why I’m maybe intervening, or maybe why I’ve given you this very specific because maybe there’s only a matter of hours before we have to do this.”

Steve Pearce: But as soon as you outline why you’re doing that, I’ve never really had any pushback from that, as long as you’re clear. It’s when it’s the wrong feedback at the wrong time in the wrong order that people are like, “Oh, I can’t deal with it.” We’ve [crosstalk 00:41:12].

Sandro: Sometimes if it’s too direct, it could also undermine the ownership that you gave the person in the first place if you’re not understanding when you’re giving feedback or if it’s at what point you’re giving the feedback.

Steve Pearce: Absolutely. I was going to say one of the things that we often joke about is when people give feedback, and it phrases a question, but actually it’s a request. In other words, have you thought about moving the button from left to the right-hand side? Just remove the question mark at the end and just say what you mean.

Sandro: Totally.

Steve Pearce: It’s training people to ask the right questions. There’s a great book, by the way, which is Leading by Asking Questions.

Sandro: Leading by Asking Questions. Okay. Noted. That’s actually an interesting thing. Also, the directness of feedback. You manage, what, 50, 60 people all over the world. How do you also give feedback, especially design feedback across cultures that probably also varies, or do you have like a process enough that that doesn’t come up much anymore? I always feel like culture is a big element of that as well.

Steve Pearce: Yes. You go through these different process and rituals that a team goes through at different levels of scale. As you get larger, obviously you have to be very clear, and I have to be very clear with my leadership team that they are accountable for driving that piece of work with their team, what an intervention would like, then I describe and go, “If it’s all going fine, you won’t hear from me.” That’s proper delegation.

Steve Pearce: If I’m still touching it, I haven’t really delegated, so they have their every right to tell me, “Back off, I’ve got this. You’ve delegated this to me. I’m responsible. You should hold me to account.” Therefore, when you’re going across different teams and different locations, it’s really important then for them to each understand what they’re accountable for, and what stream of work they really own, and how they should work together because we’re peers.

Steve Pearce: I’ll be honest with you. When it comes to geo disbursement, like we have, it is really tough. There is no way of sugar-coating it. The different cultures, the different languages, the nuance of how design is like to convey things, and it can be quite non-specific, and emotional and not very practical and so on. We’re constantly trying to improve on that.

Steve Pearce: What we have found, though, is that when people are paired up in the right ways, that’s often the quickest way to get feedback and get things moving. It’s about really clearly describing, “You’re on the hook for this, that you have this degree of accountability, this degree of autonomy. Here’s how you can make decisions. If you need anymore information, I am here. Ask me these types of questions. Otherwise, you can go fix these.”

Steve Pearce: I took the entire team through the accountability ladder when I first joined. I can talk to people about this, but I don’t know whether people would be interested, but accountability is a great framework for how we talk about holding people to account.

Sandro: Yeah, I think that’s very good advice actually. I just hear this coming up over and over that culture is an issue, but I like the framework for accountability because it can span across cultures actually. I think that’s very good advice. Also, I just saw only 10 minutes left.

Steve Pearce: Okay. Wow.

Sandro: Yeah, it goes fast. I have one last question, so we might go five minutes over, but I really wanted to ask that. You told me before that you’re going to make an announcement somewhere next week. How do you think about incremental design changes versus actually starting from scratch and rethinking a whole website, or app or experience? I see with a lot of the processes that we talked about before it’s very incremental, and we sometimes don’t take a step back from something that might get old into one direction, and maybe the times change. It sometimes maybe necessary to do a bigger change. How do you think about that? After this answer, we’re going to go into the Q&A, so please ask the questions and upload.

Steve Pearce: Thanks for the question. It’s one that I think we’ve had the most often here at Skyscanner. We have a very successful and profitable business as a metasearch company, but like we said, it does have an Achilles’ heel, which it’s input, output. Now go get it yourself on BA. Bye-bye. You’re on your own. That’s tricky because we’re very profitable on that model, and that’s very good.

Steve Pearce: In some ways, as a designer, you’re not looking to disrupt that per se, but as a business, you are very aware that if you don’t build the thing that kills you, someone else will, so you’re constantly looking at your flanks. One of things, as I described the Achilles’ heel, means that if you’re waving goodbye to every single customer, how can you really service that customer through their entire life cycle?

Steve Pearce: I know you just alluded to some things that we might be releasing, so watch this space over the next couple of months as we seek to fix that. In other words, becoming a transactional business where we will actually take the bookings ourselves. We successfully launched that in Singapore, in Australia, and that will be coming to other markets very soon, which then means that we can take care of the customer.

Steve Pearce: That, in effect, was a collective decision not just with myself, but with product, with engineering, with leadership. You start to model that out what that means for the business, and it becomes a thing that is very necessary then for the business go through. What we then did is you can carry on incrementally on the current business, but you need to be exponentially-minded on the new business model. You can take much greater risks and go much faster.

Steve Pearce: When it comes to the operational risk, the risk register, there are operational parameters where if we want to change something in the product … I don’t know. Let’s pick something very trivial. Let’s change the button color, the main CTA color on the website because we believe it will bring a bit more of a harmonious design, and it will convey our values a bit better, et cetera, all this kind of stuff that designers will do.

Steve Pearce: The business will say, “Okay, here’s your operational parameters. You can go -0.5% on revenue or GMV. If you go below that, you can’t. You have to roll back, but if it’s within those tolerances, you can proceed.” You need to establish those rules of governance for making changes even when it’s incremental. But the bottom line is you will always end up in an incremental place with certain business things that you know work, because you got to keep the business going on. It’s then to find the next things that you can be far more risk-taking to try and drive the business forward in other ways.

Sandro: You’re doing in almost separate teams basically, certain task forces almost to then work on a new product in that sense.

Steve Pearce: Yeah, in some ways that’s the easiest way to do it. You then do have certain cultural problems that arise from that which is, “Hey, why wasn’t I put on the new thing, and why am I on still on the old thing?”

Sandro: Of course, everybody wants to be on the new thing, or many people probably.

Steve Pearce: Yeah, you have to be careful on how you balance that.

Sandro: Okay. Actually, we now go into the Q&A, so people already ask a lot of questions. We start with the … Well, let’s not start with the most upvoted one with the Buzz question. Let’s do that at the end. We have some more time for the other questions. Andy Robson asks can you build a culture from the bottom-up by influencing directors to take it seriously? I guess does it have to come from top-down. You maybe mentioned when you came in it was kind of already there this strong culture. But I also know you brought about culture shift. Can you elaborate that bottom-up versus top-down?

Steve Pearce: Yeah. The simple answer to that, Andy, is absolutely. If you have a good leadership team that cares about culture, they will listen to all stratus of their business, and they must make themselves very available and create that emotional and psychological safety, so that people can say what’s on their mind without fear of repercussion. Yes, you can, and you should be able to.

Steve Pearce: There’s always going to be a meeting in the middle. There are some things where executive are operating at a different altitude to the rest of the business, and therefore will always see things in probably a longer term arc than the rest of the business, so you’re just constantly just trying to map that to what is currently known and felt in the business and taking what you might call is the main bias of the company forward on that journey. At Skyscanner it happens all the time.

Steve Pearce: We can say things, make them very apparent to leadership, and leadership listen. One of the ways in which we do that, just quickly, is we use a product called Culture Amp, and we use that every year.

Sandro: Culture Amp, you’re saying.

Steve Pearce: Culture Amp, yeah. It’s wonderful tool that surveys the entire company on many different facets of team and organizational health. That does influence the culture and the decisions that leadership take. Yes, there’s many ways to do it.

Sandro: But you got to be conscious about it, right?

Steve Pearce: Oh, 100%. Yeah.

Sandro: Next question from [Avinash 00:53:18]. Do you have a specific process to feed customer insight back to the business?

Steve Pearce: Yes, we do. I wouldn’t say it’s particularly codified, but the process at the moment is one of qualitative and quantitative. On the qualitative side that would be my team, user research, which can go and generate insights on a certain time horizon, but I would suggest that if you go to far out, that you don’t actually connect those back into the business to actually act upon in the near term, so you have to choose where in the time horizon you’re actually placing your research and insights.

Steve Pearce: It’s very important that you do that in a business of our scale with quantitative data as well, so user research and data science get put together. The way I characterize that is if quantitative is saying what is happening, the qualitative says why it happens, and you need both for the business to run.

Sandro: That actually goes already into the next question. Maybe we could talk a bit more about that. Those guys are going to use A/B test or I guess multivariate testing for a whole new features. How much quantitative data do you gather for new features versus like maybe you rely more on qualitive data? Do you have a split, or does it vary on the situation? How do you think about that whole research quantitative and qualitative? How do I have to envision that?

Steve Pearce: To answer the question, yes, we do multivariate testing. What we have learned, though, is that whilst you can test everything, not necessarily everything that you test is important, and it can slow you down, and you can end up in analysis paralysis where everything that you do doesn’t actually really change anything. It’s just another change.

Steve Pearce: What you’re really trying to do is drive business metrics better. When it comes to qualitative, I would say that that informs the concept and what the customers really care about when trying to develop a new feature, and you try and protect that at all costs because that’s effectively your NVP, if I were to use a loaded term. And then when shipping that you’re testing the variable aspects of that concept to see which bits have cause and effect that drive an improvement in the metrics.

Steve Pearce: But, like I said before, we were testing absolutely everything. We are doing less of that now because we actually want to take more of a bolder approach and a courage of our convictions that this is actually better. Naturally, the proof is in the pudding. Very simply, when we have a new feature, we will put it out to a control cohort. Those that know how to do experimentation will know what I’m talking about.

Steve Pearce: You’ll have a control. You’ll have a certain cohort. You’ll do that maybe to 1%. You’ll see what it’s like over a period of time, and then you can either ramp up to see if there are any false positives or negatives that goes in the data, et cetera. That’s all very scientific, but it can lead you into that incremental mindset, which is you got to take things on balance.

Sandro: I guess this already answers this question from Will as well a little bit. I mean, how did you go about understanding how users make choice about flights? Maybe it goes into what you mentioned at the very end on what’s going to change with Skyscanner and how did what you learn from them affect your team’s design choices. I guess it goes a little bit into that, but maybe you can elaborate a bit. How do you also prioritize what you learn then prioritize into what’s actually important?

Steve Pearce: Sorry, I’m just reading the question. Understanding how users make choices about flights, and how did what you learn from them then affect your team? Well, in effect, our results page, we call it the day view, which is your own specific days where you want to fly. That’s effectively like a new home page. That’s where all the traffic is, so you have an amazing amount of eyeballs on there. You have a lot of data that you can analyze there, what filters they’re applying, what they’re trying to do with the information, how quickly they go in, find out more information, come back out, et cetera.

Steve Pearce: It’s quite actually easy to determine what they care about. Where it becomes tricky is they actually exhibit different patterns on different platforms and in different markets. Therefore, that’s about product market fit. I’m trying to think about what would be one of the things they care about. Here’s an example. We released about a year ago an eco feature. It’s just a simple badge that when you’re searching for flights to New York, we will badge the most sustainable and ecologically-friendly flights.

Steve Pearce: That’s based on insights and emerging trends. That people are caring more and more about the environment and what to make better choices. That’s the insight. Then it was, “Well, what data do we have available that we could actually start to address this and talk to the audience?” There was a whole lot of data being crunched about that. We ship that, which was based on, “Here’s the aircraft type that flies that route. Here’s how it’s calculated and so on,” and then you’re doing a comparison. It’s basically the lesser of all the evils.

Steve Pearce: One of things that we learned from that is people were getting in touch with us very, very quickly just going, “I can’t believe you’re doing that feature. That is amazing. I can now actually make more informed decisions.” That starts to speak to our mission as a business, which is to lead the global transformation to modern and sustainable travel. That gives you one example. Well, I hope that was interesting. There were lots more slightly more boring ones, but you can see that feature in the product itself.

Sandro: Yeah, it’s very interesting. We have to slowly close here. I saw a few people are asking about your favorite resources like books, and what kind of design tools do you like, prototyping tools, and books and any other resources. Do you have some? You can just mention them, and we’ll put them in the chat for people if you have them on your mind.

Steve Pearce: Okay. Well, it’s one of my sayings. We produce a lot of flat mocks, don’t we, as designers. If a picture paints a thousand words, a prototype will save a thousand meetings. In that case, I am a big fan of Framer X.

Sandro: Framer X.

Steve Pearce: Framer X. Oh boy, that’s good stuff. Yes, of course, there’s Sketch, there’s Figma, there’s Marvel, there’s Principle, there’s InVision, blah, blah, blah. There’s so many, but I’m not afraid in saying that Framer X is my favorite.

Sandro: All right.

Steve Pearce: That’s one of the tool. We use Dropbox. We love Dropbox. We would be lost without Dropbox. Do we love Jira? We’ve got a love and hate relationship with Jira.

Sandro: There’s many.

Steve Pearce: But having said that, we do love Atlassian. We think they’re a great company. One of the resources there is their video playbooks on things like how to do a pre-mortem. They’re really lovely resources. That’s good operational governance for the team to understand all those methods.

Steve Pearce: In terms of I would say my favorite books, there’s so many. Okay, I’ll reveal my little cheat. My cheat is this. It’s a little app, and it’s called Blinkist.

Sandro: Yeah, I know Blinkist.

Steve Pearce: You may have heard of it.

Sandro: Yeah.

Steve Pearce: It enables you to read the main premise of a book without having to read the entire book. What I do is I use that to skim it, and then I’ll go and read the entire book. That’s a good way of getting the precis, the summary, et cetera, and then dig deeper into the examples.

Steve Pearce: Some of the books that I’ve read over the years that have been very helpful to me. There is Jim Collins From Good to Great. That’s very good. The one that we mentioned earlier. What other ones are there? I think Deep Work is very helpful, and then more recently I think the two books that I’ve enjoyed a lot are I think it’s Jake Knapp’s book, Making Time, and Julie Zhuo’s book, The Making of a Manager.

Steve Pearce: In fact, The Making of a Manager comes very close to home and very good examples. Of course, her writings are wonderful, so kudos to Julie. There’s so many. I’ll have a think about some more to send to the group.

Sandro: You can always share them with me, and I will share them with the audience after we’re finished, if you want to send me your reading list. Okay, so then we come to the final question. Why is Steve called Buzz? 19 upvotes.

Steve Pearce: Oh, dear. Well, it’s a real anticlimactic answer.

Sandro: That’s okay.

Steve Pearce: Back in the day when Toy Story came out. I think it was about … Someone who shall remain nameless, but who is the founder of Ustwo. I won’t make him name this. Matt Miller, Mills, at Ustwo gave me my nickname whilst we were at uni together. It stuck.

Sandro: And it stuck. It found you.

Steve Pearce: Now I embrace it. Sorry, folks, I’m sure you wanted some big elaborate story, but that’s all it is.

Sandro: Great. Thank you very much. Maybe the very last thing. Is there anything you would like to instill into the audience or leave the audience with that’s important to you, some kind of advice you would like to give to aspiring design leaders, anything like that?

Steve Pearce: Oh, that’s a big question. I don’t really want to sound I suppose like everyone else. I suppose my personal mantra, and this was passed down to me from my grandparents, and my father and so on, is a very simple saying, which is stoop to conquer. Don’t have hubris, or ego, or pride. Really seek to understand where the other person’s coming from, what they’re doing, and talk the low position to win them round and then lead them and draw them on that journey.

Steve Pearce: Don’t try and crack the whip or going competitive and argumentative and gearing for a fight. Never, in my experience, has that ever really produced good results. It hasn’t produced good, trusting relationships either.

Sandro: Well, that’s a beautiful note to leave the audience with. Where can they reach you? Should they email you? Should they follow you on Twitter? What’s your go-to?

Steve Pearce: I am on Twitter. I’m on LinkedIn. It’s Steve ‘Buzz’ Pearce. You can email me. I can’t promise that I’ll get back, but perhaps we’ll start off on DM me and follow me on Twitter. Not that I’m trying to get my numbers up or anything. It’s probably the easiest way to filter things if you don’t mind. I’ll do my best to get back to folks. It’s been a real pleasure. Thank you for hosting me.

Sandro: Wonderful. Thank you very, very much. It was very insightful, and I’m sure folks that listened to you learned a lot today. Cheers.

Steve Pearce: Have a wonderful day. Bye-bye.

Sandro: Bye, everyone.