He points to a problem in how we’ve thought about design, trained designers, and have practiced design. The great thing about designing simple products is that you can know almost everything about them: who made them, who they’re for, how they were produced, etc. But as products get more complicated, it gets harder even for a team of designers to really understand what’s going on. They get so complicated that there are lots of places design can fail.
When we go out to urban planning , that becomes even more obvious, he says. He shows Union Sq. when it was designed and how wildly NYC has grown around it. Or, at the Courtyard Marriott chain, every element of the user’s experience has been thought through. He shows a script that specifies every interaction. But you can’t anticipate everything. E.g., JetBlue is one of the best designed customer experiences and even they got it wrong a couple of winters ago.
What’s going on? It’s all about complexity. Henri Poincaré in the 19th century tried to solve the three body problem that had been set by the French govt as an open source competition. HP couldn’t solve it. It sounds like a simple problem, but it’s very hard. [BTW, there's a fascinating history of three French aristocrats hand-computing the movement of Halley's Comet, which depended on calculating the gravitational influences of multiple bodies. Can't find the ref at the moment]
Our basic ideas about design have been based on Newton, says Tim. Design assumes the ability to predict the future based on the present. We need to think more like Darwin: design as an evolutionary process. Design is more about emergence, never finished.
He presents a few principles of Darwinian design that he’s been exploring.
1. Design behaviors, not objects — the behaviors that come from our interactions with objects. If you’ve traveled on the high speed trains in Europe, there are signs urging men to be more accurate when peeing. But at Schiphol Airport, they print a fly at the right spot in the urinal; men became 80% more accurate. That’s designing behavior; the actual object doesn’t matter.
2. Design for information flow. Nicholas Christakis has looked at how networks affect behavior. Tesco uses its loyalty card — which cost them 20% of their margins — to increase sales.
3. Faster iteration = faster evolution. Viruses evolve faster than we do because they iterate faster than we do. E.g., State Farm tried out a new idea how to build relationships with the new generation. They built one storefront for this, and learned from it. “Launch to learn.”
4. Use selective emergence. This intrigues him, alathough he doesn’t know how useful it will be in design. Rather than random mutations, you choose what might be interesting and design things that get us there through many iterations. I.e., genetic algorithms. E.g., the Strandbeest walks along beaches with a hip joint unlike any in nature because the artist used genetic algorithms.
5. Take an experimental approach. I.e., testing hypotheses. Cf. Eric Ries, the Lean Startup (build, measure, learn). E.g., Ideo.org has been working on sanitation in Ghana. Where you can’t dig septic pits, Ideo has been experimenting with low cost receptacle toilets (with bio-digesters). But people didn’t want to pay for the service. So, they gave some to families and went away for three days. All the families changed their minds and said they are willing to pay for the service (which is provided by a local franchise).
6. Focus on simple rules. This comes from emergence theory. E.g., complex bird flocking patterns are based on simple rules. [Canonical example: Termite mounds.] E.g., Bi-Rite stores in SF uses simple rules: If an employee is within 10′ of a customer, you look the customer in the eye. If within 4′, you talk with them. This creates a wonderful service experience.
7. Design is never done. E.g., World of Warcraft is constantly being designed by its players.
8. The power of purpose. This creates the self-governance these complex environments succeed. Arab Spring and Occupy Wall Street are examples. Companies are experimenting with new ways of thinking about their business and products. E.g., Patagonia tells you not to buy its products because it also wants to preserve the environment.
The prototypical design artefact is a blue print. Once you created the blue print, the design was done. It was the instruction set for someone to make it. That’s how we think about design: finish and done. What replaces it: Code. It might be DNA (and Tim has people researching this), but more often it’s programming code. It’s an instruction set that can continue to evolve.
Now James Fallows [swoon] interviews him.
JF: You embody your principles. The rules are differen from a prior version. [ACK! Crash. Missed about 2 minutes]
TB: We’ve just finished designing the prototype experience for the new health care exchanges. It will affect how people choose which health care insurance to choose. Today it’s done with paper. Under the new health care laws, lots of people will get to make these choices. We worked with the CA Healthcare Foundation to prototype the user experience. What are the key pieces are parts? How can we keep the choices reasonably simple? Then each state will use this a platform to develop their own.
JF: And the govt had the wit to come to you to do this?
TB: The CA Health Care Foundation…
JF: What are the barriers? Does it cost more to do it your way?
TB: It’s often less costly. Most often they don’t have a good understanding of what their customers go through. When a health care org comes to us, relatively frequently we find out that a senior exec had to go through the health care experience. It’s true of all organizations. We don’t ask the right questions. The urgency to change is not there, and the resistance to change is always huge.
JF: Has the TSA come to you?
TB: Yes, but … well, we learned a lot. In the previous admin, we worked with them to find areas of change. Although going through the scanners has to improve, a lot of it has to do with the behavior of the people. They looked at a training program that was intended to take away some of the rule-based system they used. The more rules you apply, the less sensitive the system is. You need to give the people in that system much more independence to make judgments.
JF: Who do you hire?
TB: We look for a wide range of people. Many disciplines. We look for deep skills, and for empathy. It’s hard to solve problems for others without that. Also, most of what we do is too complex for individuals, so we work in teams, and thus people need an enthusiasm for empathy.
JF: Any unusual interview techniques?
TB: We put people into a situation in which they’re practicing design. E.g., intern program. Also, competitions. And we use Open Ideo as a way of seeing how people work.
JF: Beyond the toilet, what else are you doing for “design for poverty.”
TB: I got excited when I saw the opportunities for design in some social design work. At Open Ideo we’re working on clean water, early ed programs, etc. Ideo.org is a non-profit org. We want it to be sustainable and scalable so we look for external funding for it.
JF: How do you approach environmental sustainability?
TB: We try to build that into every project. Every project affects the environment. We try to bring sustainable thinking around systems, materials, energy flows, etc.
JF: What projects are you proudest of?
TB: The work we do in health care, including with Kaiser Permanente. Also, consumer-facing, post-crash financial services. PNC digital wallet. “Keep the change.” Etc. This is not an area where design has had much to do.
TB: For physical objects, it peaked maybe 20-30 years ago (with Apple as an exception). But we’re in ascendance for behavior-based designed. We get 25,000 apps a year for 100 openings. We’re a 600-person company. Etsy, Kickstarter, sw designed better than ever before…great things are happening. Soon if not already the number of digital designers will be greater than all other designers combined.
Q & A
Q: Your principles are so close to Buckminister Fuller’s [says the guy from the Fuller institute]. But the boundary between social and evolutionary systems is illusory.
TB: Yes, Fuller figured this out a long time ago. We’re perhaps resurrecting ideas, as every generation does. Design has operated as a priesthood for too long. When I started, I was only interested in how beautiful something is. That’s so much simpler. Opening design up to many more will convince us all that we’re all part of this big design ecosystem and have a responsibility to be thoughtful about the contributions we’re making to the world around us. I hope professional designers learn to enable that, more than controlling it. The B School at Stanford is introducing non-designers to design, which is great.
Q: What can we do to simplify the rules?
TB: The unstated bit of my thesis is that you still have to stop and design something. We develop an idea, perhaps more through iteration. That process doesn’t change. For rebuilding a complex system, maybe big data will help us to see patterns that allow us to understand what we’re designing’s complex effects…but I don’t think we’re there yet. We should be thinking about the hooks we’re building in. I’m big into APIs that allow other people to build with what you’ve built.
Q: Is it training or DNA that determines a good employee for you?
TB: Both. We hire people straight out of grad school because they’re moldable. We hire older people, but it’s harder for them to adapt. I don’t have much control as CEO. The future of all businesses is to have cultures that are a s self-governing as possible. That’s much more resilient and agile than cultures built on inflexible rule sets.
Q: I chair a land conservancy. We create parks in urban areas. Does Ideo have much experience in designing to create behaviors that will get people to use parks? What’s your view of the state of park design?
TB: We don’t have a lot of expertise in designing anything because we like designing everything. The High Line and the West Side park in NYC are remarkable examples. Projects like that show that parks can be remarkable assets to the city. We’re working with High Line on the third phase of that project. NYC’s life expectancy has gone up 3 yrs. Two explanations: People are closer to health facilities, and people walk more.
Q: What are the logistics of running a decentralized org? Mentoring? Sharing a vision?
TB: Purpose creates a sense of direction, so we talk about why the heck we’re doing what we’re doing. We think we should measure everything we do based on the impact it has on the word. We’ve done an occasionally decent job of mentoring; that can be a problem with a decentralized org. It’s a tension. Most of our employees probably want more mentoring, but we also want autonomy. We are not big believers in warehousing knowledge. Designers hate reusing other people’s ideas. It’s much better to have knowledge systems that inspire people to think in new ways. So we’re a storytelling culture. It’s a bit of an obsession of ours. If you do a piece of work, your job is to have some stories to tell about it. That’s more effective than big reports that live in a database somewhere.
(JF calls for all remaining questions)
Q: My group works with at-risk youth. Education is increasingly standards based, but your work is collaborative.
Q: How do you look at chaos? People in open markets are open and affectionate. In corporate controlled spaces, people shut down.
Q: Does form drive function or vice versa?
Q: Apple is a closed system. Google wants more control. Open vs. controlled systems?
TB: 1. University ed is not always the best way to teach entrepreneurship. Apprenticeships are interesting. 2. Great markets are vibrant, but not chaotic. I take clients to the Ferry building to point out all the interrelated pieces that make that such a great experience. It’s not top down, but you can see the patterns and use them as inspiration. 3. Form follow function? Hard to kick that notion because I believe in beautiful engineering, but most things we’re designing today have hundreds of functions, so you can’t get a single form for it. 4. I love closed systems but I think they’re inevitably part of an open system. IOS is part of an open system of everything else that I do with it. We need both. [At last! Something I disagree with! Sort of! :)]
[Fantastic. I've been a huge fan of Ideo's work, and Ideo's organizational ethos, and Tim Brown, for a long time. So I felt particularly narcissistic as I heard this talk through Cluetrain and Too Big to Know lenses. Substitute "knowledge" for "design" and you get a lot of the ideas in 2b2k. To hear them coming from Tim Brown, who is a personal idol of mine, was a self-centered thrill.]