Podcast |

Clearleft’s Andy Budd on designing for users

Andy Budd is a leading voice in the UX design community.

He co-founded the influential Brighton, England digital product and service design agency, Clearleft, more than a decade ago. Out of the office, Andy continues to push his craft forward through curating major conferences such as dConstruct, Leading Design, and UX London.

I recently caught up with Andy for a great chat. We covered a range of topics including why young startups tend to prioritize development over design, how external design expectations and internal design budgets became so misaligned, the role designers will play in the new world of voice and chat-based UIs, and much more.

If you like what you hear, check out more episodes of our podcast. You can subscribe on iTunes, Stitcher, or SoundCloud, or grab the RSS feed.

What follows is a lightly edited transcript of the episode. Short on time? Here are six quick takeaways:

  1. Companies like Slack and Dropbox aren’t successful because they were the first technology of their kind. Rather, their unique selling point is the design of their system.
  2. Hire the best designers you can afford early to solve the tricky, existential business problems. Then, bring in a more junior team who can evolve, maintain and support your product as it goes through its middle phases.
  3. Consumers expectations for digital design have risen greatly. The problem? With so few designers in budget-making positions, expectations of cost within startups haven’t risen to accommodate that.
  4. Agile development has a place in design, but companies need to balance that with longer-term systems that make design, technology and products scalable.
  5. As more startups bring design in-house, design agencies must focus on upscaling internal teams and teaching them new skills.
  6. Voice-activated interfaces might offer the appearance of natural language, but designers will still need to define their decision trees, the logic trees, and how these interfaces react.

Des Traynor: Our guest today is Andy Budd, CEO of Clearleft and a UX designer by trade. Thanks so much for joining us.

Andy Budd: It’s nice to have the opportunity to chat, because I haven’t seen you for so long.

Des: The last time we did this, the audio didn’t come through, but we got an amazing blog post out of it. It’s actually a funny story: that blog post gave rise to what Paul later coined as “the Dribbblisation of Design,” which is still one of our most popular posts.

For the sake of our listeners, could you introduce yourself and tell us a little bit about your career?

Andy: I am the managing director and a co-founder of a consultancy in Brighton, England called Clearleft. We were one of the first agencies in the UK dedicated to the field of user experience design. When we started in 2005, there were lots of people practicing usability and architecture, but none of them had really staked their claim on the UX space. We were really good friends with Adaptive Path. We saw what they were doing and we wanted to bring that slightly deeper, more considered design thinking into a realm where, at the time, the average agency would start by opening PhotoShop and moving things around on screen.

As user experience has grown up, we’ve morphed over the years into a digital product and service design agency. We help a whole range of clients – large and small, startup and traditional – tackle thorny business problems to produce some kind of new product or service.

Des: And your agency has spun out a few products in that time as well?

Andy: We have. It’s more been scratching our own itch. A lot of agencies try and deliberately diversify their income, because agency income can be a little bit bumpy at times and you can often only model three months in advance. Many have been seduced by the 37signals approach of, “Let’s move from an agency to a product company.” It’s really difficult to do both well. We never really planned to do that. What we’ve done with our products is we’ve seen things that we felt needed to be in the market and no one else was doing them, so we did them ourselves.

The first one was a product called Silverback, which is a low-cost usability testing app for Mac. Back in the day you had to use video cameras and microphones, and it was all very inefficient. We created a really small little tool, which we sell very cheaply, to democratize usability testing. It’s been really well received by the community, it’s been used by everybody from NASA to Twitter to the Obama for presidential campaign, and Intercom.

Silverback, Clearleft’s usability testing app for designers

It’s a lovely little tool, but we never set out to make it a big money business. There are other competitors charging $1,500, and that was making it difficult for small teams to test their stuff, so we wanted to put something out there that covers its development fees. I think it’s $79.

Why design is the new USP

Des: When we spoke last time, you mentioned how startups are traditionally tech-heavy. They are started by developers, they create great places for the technologists to work, but for designers in those areas, it may not be a great place, because you tend to be beholden to whatever’s being built. Designers are trying to catch up with the developers.

When you look at Stewart Butterfield at Slack or Scott Belsky at Behance, they’re all designers true and true who have created successful companies. Has the more recent trend of popular designer-founders changed any of this? Or are those founders an anomaly?

Andy: It’s definitely getting better, but there’s a good way to go. Founders usually come from a business background or they’ve got a great entrepreneurial idea. The thing they absolutely need is a developer. There’s no point being an entrepreneur and having a designer but no product that you can actually get people using. It’s relatively easy to find a designer from Dribbble and get them to come in to do a little bit of design work.

I’m still seeing lots of companies start with a bias towards business and technology. Often, it’s only when they’re two to five years in – maybe when they’ve raised a couple of rounds of funding and the demands of the fundraisers or the VCs are putting more pressure on them to release better-quality products and services – do they really start doubling down on the importance of design.

Ten years ago the competitive advantage you had with a startup was only in the technology. That was where the investment was, that was where the IP was. I think with the rise of cloud computing, libraries and platforms, and outsourcing – with the exception of artificial intelligence, which I think is becoming the new USP, the new defensible property – design is still one of the few areas where you can make a big impact.

Slack is successful not because it was the first comms tool, but because it was the comms tool that provided the best user experience. Dropbox is the same. People have delivered file sharing software and management software for years, but Dropbox did it well and did it better. The USP is in the design of the system. It takes companies a long time to realize that.

I’ve had plenty of conversations with founders where they’ve got an 80-person development team and a UX or design team of one, two or three. Because the design isn’t built into the DNA of the organization – and it definitely isn’t built into the DNA of the founders, if they come from a business and tech perspective – it can be quite difficult to know what good design looks like, who to hire, and how to bake that culture into a company that will attract the right designers.

New startups might get the product market fit for the early adopters, but they can’t produce the level of quality that’s required to make it a really consumer-friendly, high growth, commercial product. That’s still a tough place to be at the moment.

Des: It’s hard to retrofit design. You could hire literally the best designers and stick them into an 80-person development team, but the whole process, mindset, approach and office environment all needs to change to accommodate this. It’s not even just a case of teaching how to hire good designers. There’s more going on.

Andy: Absolutely. At the end of the day, founders in the first four to six years have an amazing influence over the culture of the company. Their values become the values of the company. If there’s a founder that prioritizes investment into technology over investment into design, that is going to seep through all the levels and layers of the organization. If you’ve got the CTO saying, “We need to hire three more developers,” and your UX team of one says, “I need to have a second person in design,” you’re often going to prioritize the areas you know the best. You know what capabilities three or four extra developers will give you. You might look at your website and say, “People are using it and it looks okay to me, so I don’t really get why you need another designer; we’re going to invest in this area over here”.

A lot of people don’t understand the value that design brings and still see design as stylists that make things pretty. If your product is pretty, they think that’s the problem solved. I see designers adding a huge amount of value and actually adding it very early on in the process. I advocate hiring the best designers you can afford at the start, and then once you’ve solved the really tricky, existential business problems, you can slowly let those designers go and bring in a more junior team who can evolve and maintain and support the product as it goes through its middle phases, rather than starting with a super cheap designer.

I’ve seen it happen time and time again: companies hire a really cheap designer, do a bad job, and launch. Then they’re not getting the traction they need, and six months later they realize it’s probably the design that isn’t working. They try another person that’s slightly better, and they might go through five false starts before they get a product that is really flying. They could’ve avoided all of that lost traction and lost time in the market had they just gone and got a great design team in the first place. It comes down to what you value and your understanding of the value design can bring.

Communicating the cost of design

Des: In the ten years since you started Clearleft, we’ve seen software go through another renaissance where genuinely the best designed products are starting to win, assuming they hit traction in the few of the various criteria.

The ROI of design has proven itself pretty well. Most software people use is actually impeccably designed. Expectations for design are growing, the reward is growing, but what doesn’t seem to be growing is the budget. What’s going on there?

Andy: There’s this weird cognitive dissonance. A lot of it comes down to where design is valued within an organization, versus who is making the financial decisions. Often, those two are not tied.

All the additional stuff that was once a delighter or a performance payoff element has now become a basic, fundamental requirement. It used to be the case that caring about the speed of the interface was an unusual thing. Optimizing for responsive design, mobile and tablet-based views – all of this stuff used to once be an added extra; now it’s become a fundamental. Now when you commission a digital product, there’s so much more involved in it that is expected. People’s expectations of costs haven’t risen to accommodate that.

People are in this horrific “redesign every five years” mentality. They look back at what the cost was five years ago, add 10-20% for inflation, and assume that’s going to be the budget this year, because the people who are often making these budget decisions are not savvy buyers. They don’t know how far and fast the industry’s moved on, they don’t know the new required expectations, and they don’t fully understand what the consumer market is looking for.

The budgets are often being set independently. It will be set by the finance team, without ever talking to the designers and developers to understand, “how long is this stuff going to take?”

I think it’s true internally as well. The internal team will say, “We want this thing delivered in six months,” and often that deadline has no relationship to any kind of agile story planning. What ends up happening is the design team has to try and squeeze in everything that the organization has asked for in an unrealistic deadline to meet an arbitrary goal, whether it’s hitting a financing milestone, a board meeting, or a particular client coming on board. The people who are making these decisions don’t understand what’s truly involved in creating great, world-class software.

It’s a challenge. The people who have the budget command more responsibility. In traditional organizations, that responsibility often comes to the CTO, who would have a much bigger team than the design team. It will come to the marketing team who is spending very large amounts of money on social media campaigns and advertising, and sadly, the design team often has a really low budget, so they have really low status in the pecking order and find it really difficult to demand significantly more. They’ve gotten used to making do on quite low, slim pickings. It just becomes a cyclic problem.

What’s starting to change quite a lot thanks to startups and traditional companies in California, is the people that were once the doers and the makers – the people who were front-end developers, back-end developers, designers – are slowly finding themselves in a position of authority. They are finding themselves head of design, head of technology at startups, Facebook, Twitter, etc.

A lot of people in these companies have worked their way up the ranks, and now they’re able to have those budget conversations. They’re able to say, “No, marketing needs to be in the service of design, not necessarily design in the service of marketing. We need to focus on creating a great product, and if we create a great product and we invest in a great product, then the marketing cost will be less.” The traditional companies say, “We’ve got a really mediocre product. How do we convince people to part with money? We need a big marketing campaign to make the product seem better than it actually is.”

These tables are being turned and marketing is now being more in the service of design and product. These decision makers are finding their way into organizations, they’re becoming CXOs, CDOs, VPs, and they’re having broad-level influence. It’s positive, but the rest of the world is slowly catching up. Until that happens there still will be this power dynamic that is difficult to shift.

The pros and cons of agile development

Des: When you talked about budgeting, you mentioned how people should estimate workload and you talked about things like scrums or sprints or user stories. One thing I can’t work out is if you have a love/hate or a hate/hate relationship with agile development.

Recently you mentioned that it’s excellent as an executional machine but weak in terms of strategy because of all of what you called the shearing layers that are involved. I presume, in this case, you’re talking about the difference between the decision of what market are we in and what product are we building, what job are we solving, all the way down to what are we doing this week.

It feels like there is a bridge to be built between the daily scrum on what we’re going to do today and the overall direction of the company. Is that the sort of role you think should sit with product management or is that where UX design comes in? What’s your thinking on that?

Andy: My relationship with agile is interesting. I understand where agile development came from. It’s in reaction to what was often a very old-fashioned way of delivering software, which was spec- and document-driven, having BAs do months of requirements-gathering without even even really inquiring whether it’s something the business needed. Then you built a massive document before handing it over to development, who would spend a year and a half building it, only to find that halfway through the direction change the market changed.

From my perspective, the old way of doing things was broken. I’m definitely a card-carrying member of the agile manifesto. The idea of having conversations over documentation, breaking complex tasks down into simple tasks, iterating, having flexibility of process, all of those things are really good. Unfortunately, on the basis of those foundations, a lot of agile practices have emerged which are often anything but agile. They’re often, “Recreate the demand and the desire that humans have for rigidity.” I see a lot of people going, “Oh well, we can’t possibly do that because that’s not agile” when it might be a particularly useful technique or thing to try.

The thing I push back is creating this forced dichotomy that waterfall is bad and agile is good. There are lots of people practicing a more modern form of waterfall, who are doing fantastic work. I see an awful lot of people struggling with incredibly rigid forms of waterfall, who are struggling so much they’re drowning or are doing it in a really unreformed way. It’s the dogma that I fight against.

Holding these two things up, it’s like mortal enemies. It’s not particularly useful. We’re moving into a post-agile phase, and Lean UX and lean startup is one attempt at what was broken in both worlds. The waterfall, big design, up-front approach, benefits strategic, long-term thinking and vision-making over the realities of deploying complex business software, so it tends to be a bit more long game. Because of that, it favors designers because they like to think about big systems. They like to think about the user experience, the user flow, the user journey, and creating holistic things that are consistent throughout the experience.

There’s an argument to say that developers like taking complex problems and breaking them down into their smaller chunks, and I think there’s a tendency that agile does that. Agile tends to favor a developmental mindset rather than a design mindset, and that’s a bit of an arbitrary decision. You can have great agile-thinking designers who work really well in that framework, and you can have great developers working in a big systems-thinking approach. There’s a tendency for one or the other, and that creates these frictions. That friction happens because we’re trying to solve complex problems with a very simple metaphor that we have one pace. In agile you work to the pace of a two-week sprint or month-long sprint.

Pioneers, settlers, and town planners

It’s more likely that organizations work across two or three different levels. There’s this idea of organizations basically being broken down into pioneers, settlers, and town planners. If you look at a company through that lens, the pioneers are often very agile. They’re the part of the organization that’s breaking new ground, creating new products. Those are the kind of people that are probably attracted to early stage startups, but they can only get you so far. Pioneers don’t want to put systems in place and don’t want to settle.

The settlers are the ones in large organizations saying, “We’ve built this thing and it’s all a bit sort of ramshackled, but how do we bring some structure into it? How do we create a design language that is consistent for the users? How do we start creating tools, how do we start dividing up the land so we can settle it and the pioneers can utilize it better.” Within organizations, there are always people that are looking at that systems optimization process stuff.

Then you have the town planners, the people that are going, “We’re growing really, really big. We can’t just be a town full of settlers who are going out and making their own decisions and being their own bosses. We also need to create structures and put governance in place.” These are the people who are often not thinking in terms of six-month or two-years visions, but in terms of five-year visions. There’s this context of now, next, and future. The pioneers are often working on the now, the settlers often working on the next, and the town planners are often working on the future. You need to make sure you’re working in a way that allows for all of these systems to work simultaneously.

You’ll tend to find that the product teams are really focused on the now. They’re focused on the thing they’re working on at the moment. You tend to find that the product managers, if they’re more entry-level product managers, or more entry-level UX people, they’re probably somewhere focusing between the now and the next. They’re probably focusing on creating design systems and figuring out what happens in the next block of work. The product and the UX directors are the ones who are trying to put these systems in place to make design and technology and products scalable.

This idea that it takes a village to raise a child. The same is true of large organizations. It’s the gaps between the now and next, the next and the future, the town planners and the pioneers that create friction. Often within organizations, one culture is dominant. In startups, it tends to skew towards the “let’s move fast and break things” people, and in traditional organizations, it becomes the “let’s put down roots and settle and have lots of governance strategies and policies to put in place.” You need to realize that actually all three are important for it to work.

The changing nature of design

Des: Recently we’ve seen the rise of Slack and Facebook unveil its entire bot platform. That leaves an existential question for designers: If everything’s just happening in chat bubbles, what are we here for? How do you think these things change the nature of design when you can literally have products that exist inside other products?

Andy: The nature of design is changing, and it will continue to change. I don’t for a minute think the number of designers that are required now will be the same as in the future. That doesn’t necessarily mean the role of the designer will be diminished or that the agencies are dead and there’s no future for design.

Already we’re seeing a change in the market. A lot of design activities that were once reserved for local freelancers have been absorbed into Shopify and Squarespace and a more templated approach to design. If you’re running a small B&B or coffee shop, why pay a local designer $5,000-10,000 when you can buy a Shopify theme and do it yourself for a tenth of the price? The market’s being squeezed.

Also, a lot of big companies have realized the power of design and are trying to bring design in-house. That’s reformulating the way design agencies need to engage with their clients. A lot of our work isn’t just going in and building a thing or designing a thing, but it’s upscaling their internal teams and teaching them new skills.

We’re progressing from templated systems to in-house teams to the slow dissolution of what we would know as the interface. The interface was basically a physical screen that with buttons you’d press, whereas now we’ve got natural user interfaces where you control things by gestures, voice interfaces where it’s all speech recognition, and the rise of bots where the interface is just a chat input.

Do I imagine in the future that every single service will be solely mitigated or delivered by voice or text input? Probably not. Do I see that the future for Uber is only by phoning, typing in a chat box or asking your virtual assistant to call you a taxi? Who knows, maybe in 10-20 years; I don’t see that being a case immediately.

There will be a long time when we have a big overlap of physical UIs and voice- and chat-based UIs. After all, you look around you in your home and even though we’re living in this brave new world of artificial intelligence, there are still people designing switches for buttons and lights and most of your gadgets still have a physical interface. The digital interface isn’t going away anytime soon, but I think there will be a lessening. There will be people moving into Internet of Things. A lot of my digital designer friends are now moving into physical, tangible products and I’m meeting more and more people who are interested in exploring voices and interface.

Ben Sauer, who works at Clearleft, is doing a really great talk at the moment on voice-activated interfaces. These interfaces still require design. They might be giving the seemingness of natural language, but someone still needs to define the decision trees, the logic trees, and how these interfaces react to you. I’m seeing a rise of content strategy in that instance, if the interface is a content-driven rather than button-driven. You might see a reduction in the need for graphic designers, but you would still need people that are conversational designers or interface designers, and then you start to get into the idea of designing personality. If all of these things have neutral personalities, it would be really boring. I can see people building character and character animation and great character designers coming to the fore.

We will see a small, inevitable change, but I don’t think it’s going to be a catastrophe. The people who are now at the cutting edge of designing digital interfaces will slowly move into physical interfaces and voice-based interfaces. It’s a progression.

Des: There’s still so much of the world that has to be designed, I don’t think designers need to be worrying anytime soon.

Andy: There is also room for lots of different complexity. A lot of the one-function startups that were all the rage a few years ago in Silicon Valley, a lot of those will end up stopping to be apps in and of themselves and will be intelligent agents. They’ll be plugins to voice-activated systems. Those single-use things will basically become APIs. They’ll be a really simple, dumb service that doesn’t need much interface.

We might find that they are complicated things. I’ve been playing with Amy, which is a PA of artificial intelligence, a booking engine. The interface for that is email and my Gmail calendar, and that provides the glue between the two interfaces. Those interfaces are still important.

Artificial intelligence bots, like x.ai’s Amy, have created new opportunities for UX design

There will be certain things that are just going to be too complicated to automate, at least in the short term. You might find designers moving away from these single-use apps and trying to solve some of the bigger, more complicated interface challenges. I can’t see a flight cockpit being automated solely by voice with no outputs, no buttons, no dials, no altimeters anytime soon. Even if that cockpit is self-driven there will need to be designers designing dashboard inputs to monitor those activities for a good 20 years. Frankly, trying to speculate further than about five or six years is pointless at this stage.

Des: Given how dodgy Siri can be, the idea of a pilot flying using voice control would be nothing short of terrifying.

Thanks so much for your time, Andy. It’s been wonderful to have you here.

Andy: It’s always a pleasure chatting, so yeah, I look forward to our next opportunity.