The Highs And Lows Of Building A 30-Person Lean UX Team In 1 Year

Donal O'Mahony at UXDX EMEA. Video: https://youtu.be/Xqa7CRqYRa8

Readable transcript: edited from the recording's captions for readability (fillers and false starts removed, punctuation and section headings added). Wording is the speaker's own. Timestamps are positions in the video. Names marked [?] could not be verified against the audio.

Fleetmatics, Verizon, and why this talk is about people

[00:00:06] I'm going to give you a whistle-stop tour of my past year to year and a half in Fleetmatics. So it's now "The highs and lows of building a 30-person lean UX team in one year." I'm still going to cover the lean transformation that some of you are aware of that we've gone through, CI/CD, but it's now the subtext. I thought that Brian's talk this morning from Skyscanner was amazing, and if you could imagine, we're going through a similar process, and that's the subtext. But I'm really going to focus on the people, and also what's called a UX maturity model that I've been tracking against in this company.

[00:00:42] But first, I found Fleetmatics when I joined as brilliant and understated. Bear with me, this isn't a product flow or anything like that, but not many people know about this company and the history. It started in Templeogue and it's now completely global. In those three windows over that off-licence — the Americans in Verizon, they love the fact that we started over a pub, or it's a shop basically, and it's near a pub. But that's where it started, in 2004. There were some genius ideas in there and some genius people in there, and it took off rapidly, very rapidly. It was actually one of the biggest flotations on the stock market from an Irish company perspective in the tech world, and actually the biggest in the past 17 years, still to be beaten, which was a hundred million dollars in 2012.

[00:01:28] Now, this guy is pretty important. That's Peter Mitchell and he's my boss. It's very nice to have UX reporting into the founder or co-founder. And why he's important in this talk: he's here as well, he'll be speaking at the CTO lunch, but he's a champion of user experience and he's incredibly user centered and user centric in his approach, even though he's on the engineering side. He's also CTO of Fleetmatics, but he's become the CTO within our Verizon group now, of the entire umbrella group. That's really important and that would be the theme I'm talking about, which is champions and people.

[00:02:02] Our products are very, very focused on connected vehicles, which are quite hot right now, and ultimately we're driving towards the space that will be self-driving cars. But for now we're looking at mobile workers, route planning: how do I get there, what's the best route, what should I do first, and how is my vehicle, where's my vehicle, if you're from the dispatch manager perspective. And locally we have customers like An Post, for example. That's how the redesign is going in. I'm not really going to show a ton of our designs again, I want it to be about the people, not the process and even the output. But it's getting there corner by corner. We are transforming the product, and we're moving from ultimately a very waterfall delivery, which is classic in startups — not all startups are lean — and we're moving now into a lean transformation.

The UX maturity model, and accelerating through it

[00:02:59] Hands up who's familiar with UX maturity models? Okay, so maybe about 10%. I recently did a master's in UX and I focused on this initially, and then I had to abandon it because I was actually living it, and I then focused on the actual teams themselves. But the UX maturity model is well covered, and it's this. You're starting from scratch, generally in an engineering-led organization, and it is step one, or phase one, which is unrecognized: it's not important. It's really good news that that wasn't the case in the company that I walked into, but I have worked in companies where that is the case. And again, I've really liked the themes that have gone so far about the fact that it's not last year, where we need to get along and designers need to work with engineering and engineers. It's actually now becoming about one team.

[00:03:52] This is important, this bit. When I read this in research, Nielsen Norman, that it takes 20 years to move from stage two — and there are eight stages — up to stage seven, and then another 20 years to move, on average, from seven to eight. Okay, I don't have that time, so my whole job has been to accelerate this process.

[00:04:16] Stage three: UX is very important and formal programs emerge. Apart from hiring, that has been the large part of my job. And not just my job — I'll go into building the team in a minute, and happy to talk to any of you afterwards if there are questions afterwards about how to build a team as well. But I don't come up with those programs and process, it's actually the team that does. And we do it with — it's not the UX process, it has to be the CI/CD process, or the Kanban process, it has to be cross-functional.

[00:04:46] So we're very much there, and the investment is there and committed. UX is critical and executives are actively involved. This is really where a lot of companies are floating at the moment. It doesn't say all executives, and that's why the next stages are really important. UX is one of the core tenets for the organization's strategy, really, really engaged on every level. And then the one that takes another 20 years to get to — I hope it certainly doesn't for us, I don't have the time, none of us do, and let's face it, software is changing so quickly and accelerating so quickly, I think those years are going to come down dramatically. But this is the holy grail. And it does help if your founder or that key leader is very, very design focused as well.

Joining, the Verizon acquisition, and hiring the team

[00:05:31] So I was a couple of months in, as you heard, post-HMH, and building a team, and they were very focused on UX, for six years. I had heard about Fleetmatics, I'd seen it covered in the news during the recession, and their expansion. And then this happened. I was on holidays, getting off the ferry, hadn't been able to check my phone, and I'd missed many calls from Peter. In other jobs that would be a really bad sign, but in this job it was a really good sign, so I couldn't wait to find out what it was. And it was that.

[00:06:01] Another person on my team and I talked to one another immediately and we were hugely excited about the potential, but it did affect my plans. There's a huge amount of money behind this. So from 100 million on the stock market, now you're starting to talk about billions. And look at who we are as an Irish company, and look who we're up against. So that was a really exciting moment. And I found out very quickly, because I asked some very, very direct questions, that design was very important, and design as a differentiator and a design-led organization was very important to Verizon as well. And recently, obviously, the new standing for Peter is a huge benefit to us as well.

[00:06:44] Now we're starting to hear about a design-centric organization within the group we're in, and look at the group we're in. So it's AOL, that's Yahoo, it's drone companies[?], it's Huffington Post, it's everything. This is end to end, it's from the data platform, from start to finish, from vehicles, people, content. And there's a fantastic story I don't have time to go into.

[00:07:05] At the same time all this was happening, this is now, my hiring target is within reach. A large majority of the team, actually I think all the team, are here — want to give a wave, there somewhere? So they're a fantastic bunch of people, most of them sitting together. And again this shows the level of investment that's in many, many organizations now in the area of UX. I had a lot more hair before I started trying to do this. Brian said earlier that you can't hire fast and get quality at the same time, but I beg to differ.

[00:07:40] But you can't do it without this. And I'm not saying you can't do it without me, but you certainly are in trouble if you're trying to do this when you have one junior or middleweight designer and that's it. Many companies are based on that, and they're being told what to do by the CTO or the CMO or the CEO. All of us start with that.

[00:07:57] For me, my strategy then was to put in the foundation, the intelligence, the UX foundations, and not that classic putting lipstick on a pig, you know, to put on the UI, for us to be a visual bureau for engineering. And they were all tasked with winning the trust of engineers, and I'll come back to that later. We weren't forcing a process, we weren't doing PR, we weren't doing posters, because that's just the noise. Posters will come, but it has to be connections in the canteen, in the pub, in meetings, and win trust.

[00:08:31] So you have to have the UX foundation, and then the products you saw earlier are broken into a loose group of products, and you need product leadership from a UX perspective as well. So they need to be leads slash managers. It's really a management role that they're doing, it's very important. And after that then you get some seriously senior people in again, such as principals as we call them. We're talking an average of 15, 20 years' experience each, and that takes a lot of pressure off me, so I can go home and be happy with the kids, not kick the dog if I had one.

[00:08:58] And then researchers. So instead of subjective or knee-jerk reactions, we have to base this on testing. Really interesting to be following — she's not there, but Cheryl, wherever she is — on the psychological background as well, because part of the researchers' background is in psychoanalysis or psychology, and civil engineering. Not quite the same team. But this is something that's in its infancy in Ireland and I think we need to face up to it, and I think as an industry we really start to invest in user testing and research and research questions and stuff like that.

[00:09:32] Also an aspect of content and localization. We're a global company. Ireland is representing the rest of the world at times with the American mothership, and the best way to do that is actually to practice what you preach and have that global aspect in your team. And content strategy itself, which I'm building out. A design system manager: I could do a whole talk on Angular, on the fact that we needed engineering to make the decision on our framework, not UX. That's been a six month, year-long process. We've got there, we're now at a point where that's scaling up, and warts and all, not rapidly, because it's a huge thing to change. We're moving from a monolith, as our VP of engineering calls us, and we have moved from that into a lean delivery.

[00:10:17] But what you really need now, as the next wave for the next phase, is a design system — pattern library, as it used to be called — but a living design system team manager. You might have seen the ads out for the past year. Really as soon as I got in I said I need a team manager, and we didn't even have a team yet. But I used to have the role myself, I really wanted to get that person in, and I have. I don't know what that's doing in there. Anyway, so we now need real front-end specialists.

The fantasy of UX versus the reality

[00:10:45] Anyway, awkward. So don't you hate this? This is what many engineers, product people, even UXers believe UX is. We're a good-looking team, we're not that good-looking, but we certainly don't have floating post-its. I don't know where they get them. But that's the fantasy, and it's undermining our industry, and into the software as well, that you have these designers in the corner.

[00:11:11] And this is the reality. The reality is that we're all new to this. You're new if you're the engineers — and hopefully there are some engineers that have attended this instead of their silos off in the other stage — but you're new to working with us as well. I would say a shout out to everyone who is out of their depth: just admit it. And also to new designers in the industry: please stop getting promotions every three months in UX. It's really, really bad. I can't look at any more CVs where someone's head of whatever, put in research or UX, after two or three years of a career. You need to stop doing that, or else you'll be getting nosebleeds, you'll be getting really, really worried.

[00:11:49] And you need patience. I had a lot of impatience earlier in my career as well, and I've been taught, or mentored, with the school of hard knocks previously, and with some great mentoring in the current company, that this stuff takes time. We have a mantra which is "we're in a hurry," but we also have another one which is "it's a game of inches." Both of those things can be balanced out.

[00:12:11] I was fascinated as well when the speaker this morning[?] mentioned the forming, storming, norming and performing. And I'm going to be honest, I've only formed the team, technically, to the high 20s for now, and that's only happened last week, and five people started in the past fortnight. I'm going to pause for breath. But we're really, to be honest, between storming and norming. "My process is better than yours." "The last place I worked in did this way better." "The last place I worked in, I pretended we did this better, but actually I'm realizing now I can't do this." Including myself. We all have to be honest in finding what's right for the organization and the people you're in.

Squads, workshops and concept cars

[00:12:51] I did want to focus on lean, lean transformation. We did a talk, myself and John, at Google I/O about the lean transformation, Kanban, CI/CD. But you know what, I think that's last year. I think that's something we're in now, where we're a year to two years into that transformation that Peter has led, and John as well, and I'm part of that now. But I won't go into the detail on that. If you don't know what lean UX is, if you don't know what CI/CD is, and the focus on DevOps earlier and iterating, then you're in trouble. Either you're not interested or you're probably working in the wrong organization. Controversial.

[00:13:31] But we're a squad based model. It's loosely aligned to this butterfly model, and again, to reinforce what's been said earlier, it's about autonomy. It's amazing that we kicked off today with that autonomy. So you get your North Star, your vision, from your product leadership, your UX leadership, your engineering leadership, but really it's down to the squad. You can see there, within the squads then you have UX, product owners, a greater number of engineers, obviously there's a lot to do in that area, but we're balancing out in QA. And that for us would be a mapping focused squad[?], or a data focused squad, and it could be any area specialization, similar to Spotify with the transport bar or the playlist. And they become experts and they become a gang, they become terrifying in their conviction and their knowledge, which is good.

[00:14:19] And more fantasy. I've never been at a meeting like this. It's hard, it's really tough. You hear that you're going into a squad, but you see ads from other companies that they've gone and bought some hoverboards, and it's like, what, why do I have to go and sit with this guy?

[00:14:39] That's why I'm here. I love the IT Crowd. My wife asked me, where do you work? And I said, IT Crowd. It's amazing, and that is reality. But again, please, those who are in education or those who are learning, please realize you have to work as a designer now with engineers, and they're brilliant, they're geniuses, they're creative themselves. And that's where we work, the IT Crowd, that's the reality, so there's no sad face about this. This is a reality.

[00:15:04] In that photo, Kelly here on the left in red is the only designer in any of these photos. There are product leaders, POs, but primarily they're engineers. And the ideas that come in from these co-creation workshops, from these design sprints — you don't need me up here talking about design sprints, you've well, you probably haven't read the book, but you've looked at the webpage about it and you're trying to do it. But this is the messy reality of doing it, and it's fun and they're engaged. We just need them to go and listen to the customer more, and that's all our jobs as UX designers, that you have to get engineering and you have to get product to go and visit these people, the customers, the users.

[00:15:47] This is another workshop, again a little more sanitized, a little bit more corporate, but this is not fake. This is a workshop now with customer support, with sales, and it's international. People travel around the world for this. There's an experience map in the background, or a journey, that these people worked on, because they've lived in their areas, they're experts, and they're also co-creation.

[00:16:11] So, a little bit of a pun, concept cars and connected vehicles. A real breakthrough we had was to invest a bit of time, quite a bit of time, in a concept car, and that concept car set the stall out for a vision for the next six months to a year. It was a superb opportunity to show a vision, again to excite the team. But I'd caution against doing it regularly, because really what is in there, if it lands, is that every one of these corners of the products then have to turn into a Kanban card, has to turn into weeks, not months or years, to deliver on that.

The golden ratio, KPIs and tracking outcomes

[00:16:52] When the squads are broken out you have to have a ratio. You have a ratio that says — as we called it in my bullet points for the talk — this golden ratio of how many user experience designers does it take to change a company. There is this debate about how many should you have, based on industry standards. For us the sweet spot at the moment, with teams approaching 30, to balance, for us the ratio is one to ten. And do we go as far as Intercom? Do we go as far as Booking.com, which the rumor is they're talking about one to two? Let's see what way it goes. But I do know there is a lot to build, and generally engineers build that, and we have to be very respectful of the fact that that's the cross-functional team that we're talking about here. But we certainly need to get down from the high tens.

[00:17:46] So we have a focus as well on not just KPIs. KPIs are obviously hugely important. As a team we're focused on KPIs, but as a company we are as well. So our engineering leadership, our product leadership are tracking to the same KPIs, NPS, and previously mentioned, employee NPS or internal NPS. But we also have data science in the company, we have data scientists in the company, and they're engaged in these success metrics. All that measurement is not the UX team trying to make a case against someone else or on behalf of the customer, it's doing it together.

[00:18:21] So our KPIs are all about outcomes, and again that's the lean methodology. Not about the delivery business, the product manager, the project manager telling you that we've got to deliver it at this date, but we're asking, and probably should be asking more, what are we delivering, what customer value? For models, I recommend the HEART framework. Google Ventures brought in the HEART framework and it's a very simple way. You pick out two or three of these letters — happiness, adoption, task success — and you measure against those as a team, as a cross-functional squad. You picture your metrics.

[00:18:53] So task success for a routing product would be optimized without any failure in a period, and we're tracking in the right direction. Emotional trend line from usability, for example. I'm tracking in the right direction — if Peter is still here, that's completely doctored at the end. I think one of our new designers got excited when I said it was. So it's doctored, but we are very, very open and honest about what's going well and what's not, and facing up to it.

[00:19:18] I really think that the core and the future for user experience, or whatever we end up being called — I think you're ending the day with a controversial talk about, you know, the gold rush is over, UX is dead. It's just going to get retitled. All this stuff, it's just going to get retitled, so nobody panic. But there's certainly a gold rush going on. We will die if we don't, as an industry, as user experience, track against outcomes and metrics and KPIs.

Change happens one person at a time

[00:19:48] So, rounding this out. Thanks to Kelly for this link. Thanks to Kelly Lynn Connolly[?] on my team. I was ranting about something and she gave me a great quote, which is that change happens one person at a time. And when I came in, it was all there: my connection in the interview with Peter and his leaders, and then again for everyone in my team. I tell them, look, I am very process driven, but it's actually about your personal relationships, your conversations on the corridor and in your meetings, to reinforce that it is 100 percent people and our process.

[00:20:20] I needed to study methodology[?] when I went back to academia, and all I did was then change my entire focus and my master's thesis to be actually that cross-functional team dysfunction. And warts and all, we have it, but we're owning up to it, and we're trying to solve it and we're working on it. But if we just hammered home lean, lean transformation, transformation, it's just not going to work. So it's the people that have to deliver. And to reinforce what Brian said this morning again, it's meetings, it's ownership, it's autonomy.

[00:20:54] And that takes me back to my last slide. I hope I'm ahead of time, you've got a bit of time, a battery of questions. And again, I'm going to be honest, I'm probably a little bit too honest. Wow, look, the spelling mistake, anyone spot it? Anyway.

[00:21:09] So we are here. In some respects we're not, we're a little lower, based on individuals, because it's people. Are they at the meeting when I'm talking with their arms folded? Do they think we're the arts and crafts department? Or are there people leaning forward and going, wow, that's exactly the same frustrations we have? And it's actually because it's the frustrations of the customer and the user.

[00:21:36] So the good news is, we have tracked, from one and a half years, and not solely down to me, from a two-ish in Fleetmatics, a Verizon company, up to four and a bit. And that is meant to take twenty years. I can credit people like Peter and others, but I have to credit as well the Verizon leadership, because they get it. That's why the company was acquired, and they're investing more in it. So I'm convinced that it's not going to take 20, 21, 22 years then to get to six. But I do appreciate that going from five to six is going to be a huge struggle, and that comes down to the personalities of the actual executive leadership in the company, and then the most recent hire in the engineering department, who may or may not think that we're the arts and crafts department. So thank you very much, that's me done.