What A Great Team Structure Looks Like For Building Better Products, Faster

Rasmus Landgreen at UXDX Community: Copenhagen. Video: https://youtu.be/joRqhV1P78Y

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.

Introduction

[00:00:11] I'll talk a little bit about what I think it takes to create great things, and how we do that at Nordea Markets, and the lobby [?]. Those two things. Just a quick introduction.

[00:00:40] I took a bit of education, worked at a bunch of amazing companies. There's never been a clear trajectory when looking at it. There's been some time around news and publishing, with issues that are not a lot of economics on things [?], but basically my interest has always been interesting problems, and justice, people, something wrong. Markets are on the precipice, but that's apparently where the challenges are lately.

[00:01:32] So, a bit about how we organize, how we get a common vision [?]. You already know how important a mission is to describe these things. And the last part is just quickly, how we make sure of what we build.

[00:01:48] Anyway, just quickly, just to remove confusion: Nordea Markets is a division in Nordea [?]. What basically it is, is making it possible for businesses to move currency from one account to the other, foreign currencies. You know, it's a phone wire, just going to forward to its own [?], gives them a bunch of things going on, you get dollars back and they keep a little bit of this. It's pretty straightforward, just normal.

[00:02:21] What we did was a little bit of inventing 100 different ways of doing this. So one thing is, if I only have... I send you presently, my name is Joseph, US dollars in your account [?]. But we could also do it to get the money in a week, and we'll see what the dollar rate is. Or you can get it in a month, but right now you'll agree what the dollar rate is, so I don't get any surprise. Or you can get it in a year, and maybe I'll extend it, maybe I've seen too much [?]. So all these kinds of things that help businesses work with their money.

[00:02:52] And why they need to do this is that we're looking at a complex business, and more like traditional companies. It's not tech companies, it's mostly old school companies like retail operations. They send... Ulster Bank [?] doesn't need a bunch of t-shirts back, and they use one currency, they send these t-shirts all over the world, they get bought, and all these hundreds of retail stores get banked, or two currencies, trying to figure out what to do with money while they're changing.

[00:03:31] And to the size of companies using the platform, the amount of money they can save, five workers, 42 business is massive [?]. Around close on the face of [?] on the platform, it's totally about 300 in your lines [?], and I still check that number every day, people assume. So the best point is that the amount of money they have to operate with makes a difference to the bottom line within this month.

[00:04:09] And that's also an important part of designing products for these kinds of people, because compared to everything done in interfaces, when the user makes a mistake, if a file upload fails or they send out an email, they obviously won't ask for something, it's so annoying, right? But if my user sends ten million euros the wrong direction, it's devastating. One thing is they're going to look like an idiot, but they might get fired. It's the company's money that's on the table.

[00:04:37] And at the same time we're trying to move work that has traditionally been analog, where the [?] got it all done, and now it needs to happen on a platform, and so that degree of trust with our users. Today, interestingly, they all hold some degree in economics, usually it's a master's in economics, and these are different companies, but digitally they feel like they're living in [?] basically.

[00:05:04] One thing is, it's okay, it takes a while for a company to grow to a size where you need to blend [?], so it's not companies yourself. But the other thing is, there all seems to be a [?] whether they stop with shitty systems, if you ask me. They're okay with whatever they're handed, which means that the desire to explore product, minimum common features, all these things.

[00:05:50] Thanks. Yes, so this is our platform and how we do it.

How the teams are organized

[00:06:05] And so right now, come time to worry that startups are organized one way and enterprises are organized another way. And maybe I landed in a locket [?], but it feels like how I've always been working. We have three core teams, each tackling different parts of the business. There's one that's actually building the trading services, trading products. That nice room services [?], again, it's about, there's a strategic partner to these companies: we need to help you do the right thing.

[00:06:46] So we have product owners, front end, again, too many extras, that's the business beauty. It's traders, it's all the analysts [?], it's the economic people who can fill in the gaps that I would never be able to uncover, things we use executive anything like that [?].

[00:07:05] And we have a team that's services and insights, that's focused more on all the stuff that makes the central platform: logging in, notifications, messaging services, all these hygiene things that have the same construction. Then one thing, the foundations team, which I haven't tried before, which is quite a philosophy, a little bit like Sanders [?], but there's a whole team only focused on maintaining our design library and making the components reusable, then testing them, and all the web browser, so many things. So that's the only focus, right, and it helps me.

[00:07:50] One thing that is best, that they're separate, that I'm like [?], is that also the whole release part of it, call it DevOps, which might be the wrong term, not sure about that. But since it's a bank, we're not able to actually push the button to production ourselves. We have to hand it to, I don't know what they're called, because what we're basically doing is with the pricing, currencies, so we can't do that also.

[00:08:18] Then the thing about working in a huge bank: a lot of people, which means a lot of dependencies, things like, as Laura mentioned, [?] who told you to take that, you know, where they are, which is super, super sad. But on the other hand, they have an entire in-house agency, behavioral insights, with anthropologists, researchers, interviewers, product people that I wouldn't be able to employ in my own team, but that can carry out surveys, massive surveys, across the whole of the corporate customer journey, not only the part we're interested in. And this is super, super useful, to be able to have these lots of research patterns or projects, and ones that weren't focused just on our product.

[00:09:18] So the cool thing is, everything we do, from idea to business requirements, and how the solutions for how we implement it and everything, and comes artistic [?], is done within these teams. And that's nice.

[00:09:38] We're sitting on the trading floor, which is a pretty cool experience. Like 300 traders in this cockpit environment shouting euro prices. Not as [?] as the movies, but I'm impressed. I thought, if I had to move into the corporate world, go all in. So that's what I did.

Getting the team to care

[00:09:59] And then a thing about how do we make these teams do the right things. This is my son trying VR for the first time, a roller coaster ride. I always loved this picture, because everything we do when building products ends up in front of a human being, and I think it has to spark some reaction, whether it's excitement or empowerment, whatever. But if it doesn't mean anything to anyone, then the experience will never take off.

[00:10:38] And that's for the team just as well [?], and you find it nicely with the difference between imposed wisdom and insight [?]. You have to get the team involved early on to have any chance of success. The thing is, if I'm not able to convince people why we're supposed to go right, then I have to question myself how I go into it [?]. I'm way past that point where I think I'm so brilliant [?] and everyone, those videos, they just don't get it. I think this is really the first litmus test of whether you're doing anything right.

[00:11:17] So, quick, quick, right? I think the autonomous teams, the core team, is the foundation for getting anywhere fast, and getting everyone to believe in where they're going, to be motivated. So, autonomous teams, everything is ready, then I just need to make some things that the users love, I'm in heaven. And that's why I made my domain [?].

Currency trading is insurance, not excitement

[00:11:51] And this is basically a reaction: we trade currencies [?]. And the thing is, basically nobody wants to bother with this. All our users have that core sort of working to in other systems [?]. They're using, please, most of the day, or they're using strategic planning tools or SAP things, and this thing about exchanging currency, that's just a small step on the way, some minor thing, of all the exciting things they get to do during the day.

[00:12:32] And just a note: trading currencies has nothing to do with trading stocks. Trading stocks has this excitement of winning and beating the market and doing something great, getting a great value. But trading currencies is about insurance, basically. It's about making sure that you can make a price [?], and if you always do it for a year, you're done, you're out.

[00:12:51] And for example, the Danish defense [?], they finally say, these fighter planes, that's just what we need, but they buy them for the sixties anyway [?], and over the next five years they have to pay thirteen billion or something like this. They won't pick it up front, they don't even have the money. If they wait five years then it might be thirteen and a half billion, and they don't have that money, they have to ask again. So what they need to do is make the plans for the next five years, when are the payments, and then just put it into the system, and they go about the business, because their real problem will be getting these pilot things on time, that's huge.

[00:13:28] So the amount of work in one of them, this is just like a practicality, right, to get out of the way. And which leaves us with some interesting obstacles. How do we design an engaging, immersive experience for people who basically just wound up in this? And that's an important point here, because we all have people from continued and centers [?], getting well, it's the West basically. So solving a, to go back, was it...

[00:14:04] Anyway, he said, take something seemingly boring and underserved and just walk there [?], and that's what I did. But the thing is, none of my users chose this product. They've all been handed it. They chose from a peer for different reasons, because of the relationship, because of some price or something. This is just what they have to deal with.

[00:14:25] So there's no way, even though there might be someone in profiles, it's not a super, super fan of Zendesk [?], they still have these internal investing, making it go, and there's this link that's there that I'm missing.

[00:14:38] There was something interesting about this, especially if you compare it with tools like to do your taxes, or if you have to get a new driver's license. I mean, try mentioning like the 99% of features that were also there. The thing is, you get an email that says, look for the pay taxes button, and go, and they take that last button, you get out as fast as you can, and then it's over with, and then you wait on to the next.

[00:15:05] That's the one thing, right? They think none of them asked for this. The other one is the years from 30 things throughout Europe, and soon decades basically, they've been sold, they consult the products, they don't ever change, and a lot of sales phones [?]. There having been one website that matches you with, this company has this problem, we probably have the solution. At least a natural common understanding [?].

[00:15:36] Many users have been told the same story, which then leads to, when we do the concept testing and validation, it's just incredibly hard to find common patterns, which makes it expensive and takes a long time to get to that point where you feel like we're really onto something. Where there is to solve here, from a number of customers in the world that say, I'd really need this, until the continued ready to invest in it [?]. I think that's the core thing, that every user stick, because they've been such a long, short lifetime, it's like upstate [?].

Meeting these challenges head on

[00:16:20] So, sort of the three [?] to meet these challenges head on. One thing that we met was again des métiers [?], just like a bunch of different banks that decided to be together, they put the building [?]. We have to finally done another thing.

[00:17:01] This is about getting into a meditation on the features [?]. Last year we did 100 one-hour sessions, which were either interviews or co-operation sessions or usability sessions with users. It's way more than [?] has talked to through that life, is always effective. But what we just saw, it's a new mental state for me, that everyone that's been in user testing, all this, for you and come home, there are two people saying the same thing, sounds really cool, I think we're really onto something here. But we need to stop, start by talking at one project, work, user testing it with what has happened throughout this year, at least a hundred interviews, right? Because these two guys there represent two percent of what we know.

[00:17:50] So we need to make a positive data quantitative [?], which basically, we're getting all the rooms into the same products, the same platform, the less beautiful [?]. This part, when we ask users for what they want, is not what to use it for, and that's very, very hard, to get to a level where the users actively engage with us on the products, to actors or just like further widget like that [?].

[00:18:32] That, installing a business expert directly in the scrum team, it's just a major factor, thoughts [?], because taking on this partner role, which the customers are also always asking for, it's basically telling them what they need to know. So the sun pops up [?], we simply have to take a different approach than what I want to, this ability being on [?], whereas I've always been only, let's figure out, we don't know anything, let's go ahead. Well, the only way we can take out anything is by assuming, for, that's why these trading experts help us shape the photos [?].

The three ingredients

[00:19:26] So, my Thor recipe [?], my three ingredients, as has been said a couple of times: the team structure. I don't think there's any other way than all the same [?]. It's not there to avoid the politics of, and, no, we just don't see how many little builds, or a product in the right time that actually creates value. Get the vision, this mission, as varied as possible [?].

[00:20:02] I think this is what you might end up with if you don't get this: basically tiny water holes that you might call a jar [?], but it's still just a couple of pillows waiting for one kind [?], the team telling the developers what to do. So, calling it anything else, but it's a bit of a lie.

[00:20:22] So that's one part. The other thing is that there's only a fraction of the end-user experience that will actually be dictated by my [?]. There's so much that relies on the performance, on their software, the user sitting on their whatever local time data, to be able to do all these interesting things. And I need to make sure that everyone throughout the team can make the right decision.

[00:20:54] And the last thing was also throwing a bit to my question about metrics on the user story [?]. I have to get the two started as early as possible, at the same time need to get started from there on the right thing. I've been too many times in the situation where, when we're building an MVP — I know about that, building an MVP, right? — the shape of it isn't really what us designers had imagined, it seems to be missing the most important parts. Okay, well, we'll go with it, it's ready. It gets out, and it doesn't move any numbers.

[00:21:33] So, okay, so what? No, but seriously, the business, they don't want to invest in it any more, because it didn't work the first time, so what will actually make it work? Developers, they just wasted a month at a time. It doesn't, be, get it right from the get-go, my experience has been. And that's new time invested in it, okay, so let's do that final thing, right, gives them an even less interest. And that's just sort of an eternal circle, and it's not, expect the fact that it's so critically important to go and understand the problems before starting anything. And also, it's not a waste of time going to work.

[00:22:18] Yeah, that was my [?] on team structure.