Why User Centered Design Doesn't Work (Anymore)

Yolanda Martin Olivas at UXDX Europe. Video: https://youtu.be/Ul270XURULk

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.

Ten years of designing for platforms

[00:00:00] Hello. Right, this is going to be fun. How many of you here are designers? Almost everyone. I really hope you don't lynch me after this, really hoping.

[00:00:19] Let's start. Hi, I am Yolanda. I'm the director of service and platform design at Farfetch. We are a luxury platform for retailers. I've been designing back office systems and tools for a very, very, very long time. It's been more than ten years that I have been designing for platforms.

[00:00:44] I started very ignorantly, let's say, not knowing very well what I was doing, for a content distribution platform called Reister[?]. It was just a matter of, we need to do something for a very complex system for a lot of people, and I didn't really know how to work it out. I just started working one step at a time.

[00:01:11] But over the years, working for companies like Canonical, being the platform design head for the Ubuntu operating system and working with open source operating systems and how they work as ecosystems on platforms, then I did work for Hearst Magazines. I built, during the two and a half years that I was there, the content management system and the distribution of content system for the whole of Hearst Magazines, and it has been used as well for Hearst News and Hearst Television.

[00:01:44] After that I also worked for the Open Bank Project for Banco Santander, for about a year, just doing the same thing: trying to get the platform for banking sorted for their digital banking products. And now just working for Farfetch, which started as a marketplace, started just as a marketplace for luxury boutiques and retailers, and over the last three years is going really far. Stores first, becoming a platform, and now becoming an ecosystem.

User centered design is obsolete

[00:02:18] But the thing is that during all this time, one thing I have learned working with platforms and ecosystems, and if I can do it, is that user centered design is obsolete. I know that all of you are firm believers of user centered design, and so am I, but it's just not enough. It's just not enough.

[00:02:45] We cannot continue designing just for the end user. We have to start looking a lot more into the effects of our products, the effects of our design. We have to look at the non-users of our product who get impacted by them, at the support staff, the customer service teams, but also at the whole ecosystem where your product lives. We have a responsibility as designers to do so.

[00:03:15] There has been a tremendous change in the world. You have all noticed it, it is great. We cannot continue having the selfish decisions that we have been making until now. We have to look at the world as a complete entity. We cannot just be basing all of our design in the individualistic social political agenda that we have had for the last 50 years. Everything we do has an impact on the environment, on the psyche of everyone, on the socio-economic models. We need to be a lot more conscious of where our products and services live and what impact they have.

[00:04:05] The truth is, during the last 30 years we as designers have all been indoctrinated in user centered design. Focus on the user, design for the user, look at the smallest slice possible and focus on it. Focus on the MVP, focus on the feature, focus on the wireframe, focus on the screen, just that one thing. And ship it fast, do not look too much around. Let's just focus on that thing that we need to do to get the product out.

[00:04:41] But that's a very myopic point of view. It gives us great focus to develop that product faster, but that user centered design has limitations. It has great strengths when we actually have to focus on a point, but it has huge limitations when we need to be looking at much bigger systems, especially when we have to look at problems like this.

[00:05:12] I believe, and I think it's been great, as we have been so many years in this profession, I have seen the evolution of designers and product design. 30 years ago the profession mostly didn't exist, and we have gone evolving and designing incredible products and services, digital and analog, all around the world, that have had incredible success. It has given us this increased power, let's say. Now designers sit at the table, on boards. We have chief experience officers and chief design officers. We have a lot more voice at a higher level.

[00:05:53] However, I don't think that as designers we have ever tried to use our processes or products or methodologies to try to solve the really systemic problems that we have in the world now. I know that it really looks big, but the truth is, do you really think that your products are not actually making these things worse? Have you ever thought about it? Has anybody in big companies thought about the huge impact the products have? We have all seen what social media is doing to teenagers' mindsets.

Uber, Lyft and coffee pods

[00:06:38] I have a couple of examples. Uber and Lyft: fantastic design teams, fantastic design methodologies, we all read their blogs, great products, amazing service. They solved a very user centered design problem, a user problem. We all wanted to get cars fast. They were going to solve the car sharing problem. If everybody took a car and we shared it, then there would be less traffic. We will ditch our own cars and start sharing rides with other people, therefore the traffic would reduce and it will be fantastic. So it was killing two birds with one stone.

[00:07:21] That is not what happened. Car sharing apps have created a worse problem than the one we had before. It was a completely unexpected result: the fact that the traffic has got worse, and public transport, which is so important to reduce carbon emissions, is being ditched because it is so easy. You usually ride a bicycle to go to work every morning, it is draining a little bit, and how easy it is to go, I'm just going to call an Uber or Lyft.

[00:08:00] Nobody really shares the Ubers. How many of you actually here share Ubers? A few, a few hands, not really. It was created to be sharing cars, but nobody does it. It's such a big problem that transport organizations are starting to put rules and trying to govern around it, because it has increased the CO2, it has increased the traffic, it has made the lives of the non-users and the users of Ubers in the city centre worse.

[00:08:44] So now governments and officials are trying to regulate around it. They're trying to think of laws to make the cars used for car sharing into electrical cars, to at least reduce the carbon emissions. So from trying to solve a user problem to creating a massive ecosystem one.

[00:09:11] That's not the only one. This is one of my favorites: coffee pods. Who doesn't love them? Who doesn't use them every day? How many of you use Nespresso or whatever, Dolce Gusto, coffee pods? Boom, that's actually quite a few. Not so many people, I expected a lot more. I've done a similar talk a few months back and it was at least half of the room, so I'm glad to say that not so many people.

[00:09:41] However, the use of coffee pods increases almost 50 percent year on year. It's tons and tons of coffee consumed like this. What was the user problem that we were trying to solve with this? It was always: make the life of the user convenient. No having to wash coffee machines, serving the right amount of coffee at the right temperature, so quick and convenient, all the different flavors that you can ever think of. If you have a party you can have everybody actually using a different pod for a different flavor. Great, isn't it?

[00:10:17] Well, it has created such a huge environmental problem. These pods are not easily recyclable, and the ones that you see in the picture are the aluminum ones of Nespresso. Even these ones, because of the glues and the aluminum on top of them, they can only be recycled in Nespresso's own facilities. But there are tons of them made of plastic, and the fact that we were trying to solve a user problem just created a huge amount of garbage that now we have to deal with. Whereas actually making yourself a coffee from a coffee bag and a coffee machine produces zero garbage.

[00:10:57] So again the problem is big, and the governments and institutions and organizations are trying to curb it. The city of Hamburg has forbidden the use of coffee pod machines in public buildings, and it's not the only one. Many more cities are starting to forbid the use of coffee pods in public spaces.

Apps that solve a user problem and create another

[00:11:22] There are a lot more examples, and if you approach me later I can tell you a million of them. One of my favorite ones, or one of the ones that creates me more anxiety, is app design, and I bet a lot of you here are app designers. I have an Oral-B toothbrush and it came connected through bluetooth to my phone, and my phone tells me for how many minutes I have to brush my teeth.

[00:11:51] I tried to use it and I thought it was fun, and it gave me a terrible amount of anxiety in the mornings, because if I didn't have time to do the two and a half minutes that the toothbrushing thing wanted, I was feeling bad for the rest of the morning. So it was trying to solve exactly what user problem? Who doesn't know how long you need to brush your teeth for? Do we really need to create an app to tell me for how long I need to brush my teeth? Not really, and it created me anxiety. I don't know about anybody else, maybe I'm weird, I'm sorry.

[00:12:23] But Instagram is creating anxiety in teenagers. Facebook is allowing for bullying and bigotry and political discourse of a negative kind to happen, and so on and so forth. So a lot of apps and designs trying to solve something kind of temporary and user centric is creating a lot of issues.

Designing for interconnectedness

[00:12:50] But I don't want to depress you, and there is a solution to this problem. What can we do? What can we all do? We can look at the big picture. It's a lot more simple than you think, because user centered design is myopic. It looks at the issue from one point of view, in one moment in time. What we can do is actually take a step back and look at the picture from a much higher perspective and start working for interconnectedness.

[00:13:32] I always have difficulty to pronounce the word and define it, so this is my favorite definition. I don't have many in the presentation, so don't worry. Interconnectedness is about the reduction of friction between interactions at larger scale systems. This helps us move beyond the concept of a user and embrace the concept of relationships and interactions between peers, and the ecosystem as the key element of a business.

[00:14:04] Basically what it is saying is, don't look just at the user, look at the relationships in between all the users in your platform. Look for that place in between where the value gets created, where the transactions happen, and what happens in between the people doing the transactions and everybody else within the ecosystem.

[00:14:28] I could go on and on about designing for interconnectedness and platform design, but the best way of learning more about this is just going to that site. It is open source and you can get all the information, the tools and the templates that you need. That's how I learned, and those are the tools that I use to practice ecosystem design, so knock yourself out. And collaborate, it is a collaborative platform, by the way.

How Farfetch works as an ecosystem

[00:15:01] So how do we practice ecosystem design at Farfetch? First, a little bit about Farfetch. How many of you know what Farfetch is? Not so many people, so I'll explain. Farfetch is a platform to help boutiques and brands in the luxury space to do ecommerce. Simple, right? We're just like an Amazon for luxury brands. We provide the tools and the services that allow all of these companies to do their everyday retail services, the retail business.

[00:15:43] And how are we actually working as an ecosystem? It just works like in every ecosystem. I was hearing the talks yesterday, being very, very inspired by Alan and others, when I heard that there have been so many talks these days about ecosystems and platforms. Something that Alan said yesterday about us all being gardeners: gardeners and gardens are one of my favorite metaphors for ecosystems, and it's true, we are gardeners. But at Farfetch we use the city metaphor for explaining ecosystems.

[00:16:19] So the infrastructure of the platform is like the urbanistic design of any city. It has the technology and the services that connect everything together, but if you look at the city it would be the plumbing, the electricity, the road design, the lighting and so on. That is what you put the next layer on top of.

[00:16:44] Then you put the businesses, in this case the websites and applications and the participants used to create and consume. That layer is everything that you put on top of that infrastructure. It can be anything, it can be tools for logistics or production or product management, or a million things, translation services, every type of business: small boutiques, multi-branded boutiques, brands themselves, as well as startups, apps, games[?]. Tons and tons of different businesses that exchange value in between them every day.

[00:17:24] None of that would work if it was not because of the community. Everyone operating within the ecosystem is what makes the city move. No ecosystem will work with just two or one element of this. Ecosystems work because there is an infrastructure, there is something that supports the businesses and the services that people use. Having services and some people with no infrastructure wouldn't succeed. Having businesses and infrastructure with no people wouldn't succeed.

[00:18:03] So looking exclusively at the users without looking at the impact and the interactions in between all the elements of this, it's just not going to work. At Farfetch we have different levels, but this is how every platform and every ecosystem works, and I'm sure you're already working out in your heads your own businesses and how they fit this type of model. You're starting to see the layers of your company.

[00:18:36] In our case, there is a business ecosystem where we put everything. All the different elements are there to create value for everyone, and that's the business ecosystem. Then we have the marketplace platform, but that's not the only platform Farfetch has. We have a data platform, we have a logistics platform, we have a production platform where we take the pictures of every product that we display on our websites. So there are many platforms in the ecosystem.

[00:19:10] The next layer is multi-channel services, and we have tons of them. Anything from supply chain tools to the WeChat[?] app. So we go from the back-office kind of supply tools to the consumer app, in this case China, but we have a Discover app on Android and iOS, and we have websites, and we have tons and tons of services on that layer. And then we have the single touch point where our users interact with those layers, and everything goes from generic to the user. But there you can see how that touch point has interactions and interconnectedness with all the other layers.

Actors and entities instead of personas

[00:20:14] Here is where actually it was one of the most difficult things to explain at Farfetch when I started. When I talked with my team then, which was relatively small, I told them: actually, I want you to forget about personas. We're not going to do personas ever again. We're not going to think like that until we figure out who are the entities and actors of our ecosystem.

[00:20:42] There was a generic gasp, because of course that's what we have learned. You do user centered design, therefore you do personas. However, when you're working with huge ecosystems and big platforms, you cannot singularize so much into a persona, because it makes the model impossible.

[00:21:02] What we use is actors, or entities, and these ones are the entities of the Farfetch platform, but pretty much they fit any platform or any ecosystem. We have producers, consumers, aggregators and governors, and this happens, I mean, I've studied and we have modelled many, many platforms on this, and it is the same over and over.

[00:21:33] In our case, on the producer side, we have merchants who supply luxury products, and service providers that supply services to the participants. In consumers we have the ones that purchase products and services, not only products but also services, so our back-office consumers are consumers too. In aggregators we have the application owners, the ones that connect the supply with the demand, again working with the intermediaries[?]; they are the glue.

[00:22:07] And the most important part of any platform ecosystem is the governors. We have at the end the stakeholders, the ones that input into the platform a strategy. We all know very well about stakeholders. In the case of big ecosystems it can also be a part of your business partners and so on, can be governments and legislators, for the case that I was saying before about governments and organizations trying to regulate ecosystems.

[00:22:35] But the most important part of any ecosystem is the shapers. Every one of you in this room are shapers of your ecosystem: designers, product owners, product managers, engineers, architects. Everyone in the company shapes the ecosystem equally. As you can see, there is no hierarchy in these actors of the ecosystem. Everyone is at the same level, everyone influences at the same level, no one is more important than the other, and that's how ecosystems work. If you drop any of them, if any of them doesn't become an active participant of the ecosystem, it just dies or doesn't succeed.

Service design as the way to practice it

[00:23:28] Because designing platform ecosystems is quite a strategic job, it is not that everything we do at Farfetch is designing ecosystems. What we use to practice ecosystem design at company level is service design. So this is how the hierarchy of working with platforms looks like for us. We have the whole platform ecosystem surrounding everything we do, but then we work as service designers, and I have several in the team. Service designers work at the service level, looking at things like acquiring partners, or onboarding, or manage catalog and so on.

[00:24:14] And then of course we have UX teams and visual design teams that work at how those services get manifested on the user experience points. In the case of acquiring partners it could be marketing or business development; in the case of onboarding partners, it's tools or documentation and so on and so forth, logistics, data, applications, reports. We never look at it just from the tool point of view or the interface point of view. We look at it as well as how we can provide that experience from every service point of view.

[00:24:51] And as I was saying, service design is what we use to drive that operational change, because implementing and bringing platform thinking to Farfetch and to any organization is really hard. It is not an easy thing to do. You don't always get stakeholder approval or board approval to do this. I was lucky in the case of Farfetch, because they knew they were a platform and they wanted to bring this discipline to the company. But still, the difference in between wanting to bring this discipline and actually bringing it every day is not that easy.

[00:25:28] So what we're doing at Farfetch is use service design to work as that intermediate point, to bring platform design to a lower level and make it a lot easier to practice. Both disciplines are very much interlinked, but that's the way we differentiate the user experience, the user design, from the rest of the platform design, by applying service design.

Simplifying the complex

[00:26:08] Now, designing for ecosystems: the main goal, apart from reducing transactional cost, which is the main business goal of an ecosystem, is making sure that that interconnectedness, that value exchange, happens at the minimum cost possible. But that cost is not always just about monetary cost. It can be friction for the users, or it can be footprint. At Farfetch we have a mandate to try to reduce as much as possible the footprint of all the transactions that we do.

[00:26:43] But that can be reduced to: designing ecosystems is about simplifying the complex. It's just making really complex systems into easy models. I'm going to show you now how Farfetch looked before we started to work on ecosystem design.

[00:27:05] So we have boutiques, we have tools that allow us to do business with the boutique, and companies that support our logistics, payments, ticketing services, customer service and so on. We call these tenants, so it's companies that do their business through us but do not participate in the marketplace. We also have companies like Store of the Future, business units within Farfetch that create their own businesses. In the case of Store of the Future, it is our IoT branch, it is what brings online and offline into the same bucket. And we also have a whole bunch of startups and third-party companies that provide apps and services for us.

[00:28:00] This is actually a year ago; now it is probably twice the size of this, because we have a lot more businesses and things, we have acquired a couple of big, big companies. This is how we look, and trying to work the relationships and the value exchange in between all of these things looking like this was pretty much impossible. It took a lot of time and a lot of effort to reduce it to this. It was months. This is probably version 23 of this diagram. At the beginning with circles and things all over the page, it took a long time and a lot of effort of applying things like value exchange profiles and entity profiles and systemic design to end up in this, which is basically, as you can see here, the actors, the services, the API, shapers and stakeholders, and how they relate to each other.

[00:29:09] This diagram, together with the whole concept of platform and platform ecosystems, a team of ten people, not only from my team but from product management, architecture, engineering and so on, we took it into something that we called last year the platform roadshow. Because not everyone in the company knew that we were a platform, or how to become platform shapers. So we did a massive educational piece and we took this diagram together with booklets and information and that metaphor of the city to everyone in the company. So about 3,000 people got educated into platforms.

[00:29:53] Everyone in Farfetch knows what part they play in the ecosystem. They understand how their jobs impact every layer of that luxury retailers' ecosystem, and that was extremely important, because as a design team we were banging against walls constantly. The teams didn't understand what we were trying to do, why we were looking so far away. If we were about to design a feature, why did we want to look at the whole service? Why we wanted to look at where the service was living, why entities, why are you doing all of this?

[00:30:31] Once we did this educational piece and we took it to everyone, from marketing teams to localization teams to everyone, absolutely everyone, those fears kind of dissipated. Everybody understood what role they were playing and everything became so much easier. I mean, we haven't won, believe me, we still have a lot of education to do and it is not an easy ride, but we are doing a lot of leaps and bounds compared to last year.

Summary points

[00:31:01] And just one last way of how we practice it. In practice it means that we divide our team, the platform team, into three areas: the strategy, the services and the touch points. We still practice user centered design at that level, we do, so it is still useful. I was not trying to say forget about user centered design practices. Of course we have to still look at the user, but not only. You cannot only focus on the one user at one time, you always have to try to look beyond.

[00:31:46] So these are my summary points. Please look beyond the user. Even if your mandate is to design one screen or one feature, try to understand where that is coming from, what is the interconnectedness that is happening, what is the value exchange, where the value chain is happening.

[00:32:06] Think of the long-term impact of your design. This is very important. We are responsible for the designs we make. What if you were asked to design the atomic bomb, would you do it if you knew what was going to happen with it? Try to think what impact your design is going to have at every level.

[00:32:29] Always try to reduce the friction, but not only the user friction: environmental friction, the socio-economical friction, the psychological friction. Be responsible. Thank you very much.