Selective

Emergence

Design
Behaviour

Fitness

Information
Flow

Faster
Iterations

Design
is never
done

From Blueprint to
Genetic Code:

by
Tim Brown

The Merits of an
Evolutionary Approach
to Design
Like everything else in life, design is evolving, forcing us
to give up the very essence of the Newtonian notion
of design: the blueprint.

LIKE MOST DESIGNERS, I am quite comfortable with the notion of

designing simple things. I can pick up just about any object and
tell you how it was made, and I could probably have a reasonable
crack at designing an equivalent of it, even though Im not a particularly technical person. Thats because it is possible to defini-

tively know everything important about a simple object: we can

know its form, the market for it, and the best method of manufacturing it. We might even know what to do with it when people are
finished using it. The traditional design process entails figuring
all of this out beforehand and making it so in the world.
Rotman Magazine Spring 2012 / 17

Unfortunately, as the worlds problems have become more

complex, the traditional design process has been challenged.
Many of the products and services created today are more like
complex systems than simple objects; they often involve a confluence of software, hardware, and human behaviour. And as systems evolve, so must our knowledge of systems design.
Take urban planning, for instance. Look around your city
and you will see countless examples of the unintended consequences of a failure to understand a complex system. Thats
because a traditional design process was likely used to create
the sub-par products and services, and cities are like organisms
they evolve. It is very difficult to create a planned city, as Chinese leaders are currently finding out through their attempts to
create fully planned, top-down, large-scale cities.
In traditional attempts to design a service, we script the
service, creating a user experience blueprint that attempts to
describe everything that will happen to the customer during the
experience. For a hotel, for instance, this would include everything from what the lobby looks like to what the check in service
is like. Attention to all these details leads to a relatively complicated script, which makes us confident that we have covered all
the bases. The problem is, even when we get these scripts right,
its amazing how often things go wrong.
From Newton to Darwin

As designers become more involved in solving the worlds wicked problems, an ability to deal with complexity becomes all the
more important. In my view, this indicates a paradigm shift for
the world of design, because it demands a shift from thinking
about the world in the way that Sir Isaac Newton encouraged us
to think about it, to the way Sir Charles Darwin thought about
it. Let me explain.
Newtons world was based on the assumption that we have
an ability to predict the world based on actions in the present.
When we think this way, it encourages us to be top-down in our
activities, to be predictive, to believe that we can imagine a complete system. I would argue that the complexity we often face today requires us to think more like Darwin, who encouraged us to
think about constant evolution, emergent change, and the notion
of unpredictability on a large scale, even if we understand things
on a small scale.
As designers and as leaders, I believe that we need to start
emulating Darwin a bit more and to stop emulating Newton. Following are some possible aspects of a more Darwinian approach
to design.
18 / Rotman Magazine Spring 2012

1. We should give up on the idea of designing objects and

think instead about designing behaviours.

Behaviours are about the interrelationships between people and

the objects that exist in the world around us. To illustrate the difference between designing an object and designing a behaviour,
take a look at this image:

This sign can be found on trains in Europe a vain attempt to

encourage the male of the species to create less mess in the public
toilets. But because it is simply an instruction, it doesnt work very
well; anyone who has used a train toilet knows all too well the degree to which it doesnt work. About 25 years ago, the fellow who
managed the airport in Amsterdam had a much better idea:

He realized that if you printed a fly on the urinal in just the

right spot you would give people something to aim at. And when
you give a man something to aim at, he actually does a remarkably
good job: this approach has reduced the mess in urinals by up to 80
per cent. Thats what it means to design a behaviour.

2. We need to think more about how information flows.

If you talk to people at the Santa Fe Institute, or read any of their

books, you will learn that a key characteristic of a complex system is that the more complex a system is, the more information
flows through it. If this is true, then we ought to be thinking more
about these information flows when we are designing complex
systems. In fact, before we work on designing a better solution,
we need to get better at understanding the complex system as it
is today, and what information is already flowing through it.
For example, Harvards Nicholas Christakis has studied
the relationships between people with respect to their health, and
one of the conclusions he has come to is that if you are in a network of obese people, you are three times more likely to be obese
yourself. Conversely, if you are in a network of non-obese people,
you are three times more likely to not be obese. This is a very important insight for design: that the behaviour of those around us
significantly effects our behaviour. Intuitively we might know
this, but we dont necessarily always think about it when were
designing systems.

ciples perhaps, but there is no guiding intelligence. We humans

have the benefit of potentially using the best of both when we
design something.
There are technologies out there that are already doing
this enabling us to use the idea of selective emergence to rapidly iterate things while nudging and guiding them towards some
outcome that we want. For example, genetic algorithms. In software today it is quite common to build algorithms that reproduce
themselves; certain rules are applied, but you dont know in advance what the optimal version of that piece of code is going to
be. Evolution gets you to it, and all youve done is apply certain
rules to nudge it in the right direction. This idea can be applied in
fields as varied as engineering, design and art.
The strange looking thing pictured below is called a strandbeest. These remarkable structures were created by Dutch sculp-

3. We must recognize that faster evolution is based on faster

iteration.

The faster we do things, the faster we learn and the faster we

improve. The natural world deals with this truth very well. The
reason viruses evolve so fast is that they reproduce every few
minutes, which is why we have such a hard time keeping up with
them. In contrast, humans evolve (i.e. reproduce) every 20 years
or so, and in general, business is more like humans than it is like
a virus. How can we make a business more like a virus? One way
to move in this direction is to accept that we cannot know all the
answers before we do things.
We recently worked with State Farm Insurance in the U.S.
to launch Next Door, which provides coaching and financial advising to a new generation of customers. Next Door is a place
for customers to come and learn about insurance and financial
services, and at the same time its a place for State Farm to learn
about them. The idea that we can launch things simply to learn
from them is quite useful when were thinking about increasing
the reproductive pace of iteration in business.
4. We must embrace selective emergence.

So far, natural biological systems appear to be way ahead of us

in dealing with complexity, but we do have one advantage over
them: with biological systems, all of the improvements are random they are based on mutation. There are some guiding prin-

tor Theo Jansen, and you can find them walking up and down
the beaches in Holland. Made out of PVC plumbing pipe, they are
self-articulating they move on their own. Jansen used a genetic
algorithm to create the hip and foot joints. He didnt know in
advance what the right ultimate solution was, so he designed an
algorithm and it kept iterating and iterating until it created the
most efficient foot and hip mechanism to make these sculptures
walk on their own. I believe we should be using this approach
more often in design. We are already seeing it done in architecture, where architects use a similar kind of technology to create
the facades of buildings the ones you see that often look much
more organic than a traditional structure.
As designers, we need to remember our relationship with
science a little bit more. Were often very good at exploration
Rotman Magazine Spring 2012 / 19

or divergence asking questions without any real sense of a hypothesis. But I think we also need to relearn some good scientific
methodology. Doing so will enable us to ask more of the right
questions, come up with better hypotheses, design effective experiments and most importantly, share our learnings.
5. We need to focus on fitness.

Biological systems naturally focus on fitness; at its core, thats what

evolution is all about striving for fitness, whatever the environmental context might be. All kinds of biological systems do this;
but what is the equivalent of fitness in business and in design?
I believe one way of thinking about fitness in the organizational realm is the concept of purpose. Organizations that have a
clear purpose tend to be able to design in a less top-down way.
Many years before Apple took over the world, Steve Jobs spoke to
his people about being insanely great, telling them: What you
create has got to be so good that you are shocked that you could
actually create something that good. This became the driving
purpose of the organization, and it allowed many more people
to contribute to the greatness of Apple than would have been the
case if there were no such purpose. Imagine an organization that
didnt have this purpose, but still had a Steve Jobs in it; it would
have been impossible for him to achieve what he achieved. I believe that his incredible vision plus an organization that believed
in the notion of being insanely great are the keys to Apples success.
6. We must accept the fact that design is never done.

In the architectural world, there is a notion of life after the open

house. Architects see all sorts of perfect photographs of buildings just at the moment when they hand it over to the client, but
very rarely does anyone see photographs of what happens afterwards. I think this is natural, and I do the same thing: I design a
product and I take a perfect picture of it, before the manufacturers get their hands on it, never mind the user. This is that moment when the thing is closest to my vision, and its when I think
Im done with it. Of course, this is a ridiculous notion, because in
truth, it is now in the hands of users, where it will be adapted and
used for things that I didnt expect it to be used for.
In the world of video games, particularly in online games
like World of Warcraft, design is going on all the time. Yes, there
is an environment that was created, but the design of behaviour
of the events, the characters all of this is done by users who are,
in effect, participating in the design. I believe that design is going
to look much more like this in the future, rather than the idea of
unveiling a pristine thing that we hand over to people.
20 / Rotman Magazine Spring 2012

Evolutionary Organizations

Embracing a more evolutionary approach to design can do more

than enable us to create better products and services: I believe
that we can use these ideas and principles to tackle some of the
most important challenges of our time. In the field of chronic
health problems, for instance, were beginning to see opportunities to hand control over to patients and enable us to design behaviours for ourselves that help to manage chronic disease. As a
result, were starting to get past the idea that our bodies are mysterious black boxes. For example, there is a scale that comes out
of Europe that does something very simple: each time you weigh
yourself, it sends the data to your iPhone. Over time, you can
build up a clear picture of the relationship between your behaviour and your weight, because you get to see it on a graph. Such
devices could measure all sorts of things, potentially allowing us
to equip people with the means to change their behaviour.
Another way of changing behaviour is to put the tools of
design themselves into the hands of people delivering services.
Weve been working for several years with Kaiser Permanente,
teaching nurses and doctors and technicians how to use design
thinking to improve patient care, and Kaiser now has its own
consulting group made up of nurses who have become expert at
this. They go around to hospitals working on different problems,
creating wards and hospitals of the future. Already, theyve had
some great success in increasing the efficiency of shift changes
for nurses.
At IDEO we are particularly interested in applying design to
issues of poverty. While its still early days, we are finding that
the way to have the most impact is to give people tools rather
than designing an end product. For example, weve been working for some time now on sanitation issues in Ghana, designing
a low-cost toilet for peoples homes. The problem is, even if you
design the lowest-cost toilet possible in this case, about $30
its still too expensive for most families. We looked at things like
micro-finance to deliver the service, but what we found is that the
really important design unit here was a service business model.
The model allows room for entrepreneurs to set up service businesses where they rent toilets out to people and come to empty
and clean them on a daily or weekly basis. The lesson we learned:
the important part wasnt the design of the toilet itself, but the
whole business system around it.
We recently worked on a project with the Bill and Melinda
Gates Foundation to create a Human-Centered Design Tool
Kit, which is really a field guide for design for not-for-profits and
NGOs. It includes tools that enable them to understand their users

If you want to be involved in the participation economy,

our old notions of designing in a controlled way just dont
make sense anymore.

better, to create ideas through prototyping, business model design,

etc., and weve had some interesting success with it. Already, the
tool kit has been downloaded over 70,000 times and it is being
used on projects ranging from creating a new maternal hospital in
Nepal to working with a cooperative of weavers in Rwanda to water distribution management systems in Malawi and hand-washing stations in Vietnam.
In an effort to scale all of these efforts up, last September we
launched IDEO.org, which is a sister organization to IDEO and
a not-for-profit design organization that works on social innovation initiatives. Unlike IDEOs core business where were hired
to solve problems for clients by designing new services or products were doing something quite different with IDEO.org: were
looking to build out the capacity for design in the social sector.
This is a space where giving up control is necessary: if you want
to participate and be involved in social networks what I think
of as the participation economy our old notions of designing
everything in a controlled way just dont make sense anymore.
Evolving as an organization means listening to not only your
clients, but to whats happening in the world around you. Our clients would often say to us, You talk about how we should use
innovation to disrupt ourselves, but how are you disrupting yourselves? We thought about this for some time, and asked questions like, What if we could solve problems collaboratively via
a global network? Basically, what would happen if IDEO had
50,000 people working on design challenges instead of the few
hundred working within our walls?
We came up with Open IDEO, which is a platform for tackling social innovation problems. We do this in conjunction with
various non- profits and sometimes with companies, as well. For
instance, we teamed up the World Wildlife Fund and Sony on
one challenge, and Unilever is helping us out on the sanitation
challenge in Ghana. At the moment we have about 25,000 people
in the design community working on projects in 178 countries.

We are already seeing businesses getting launched through Open

IDEO, and were seeing some really interesting ideas going back
into the organizations that sponsor projects. But more importantly, were learning a huge amount about how collaborative design
might happen in the future.
In closing

Like everything else in life, design is evolving, and in many ways

we are being forced to give up the very essence of the Newtonian
notion of design: the blueprint, which personifies control and defining every outcome of the design process.
Instead of focusing on designing blueprints, I believe there
is something that better represents what we should be designing
going forward: genetic code. At one level, genetic code represents
the biological view of design I have described, because it is an instruction set for biological behaviour. But more importantly, it
represents the idea that code is only the beginning of something:
it sets off a series of behaviours, but you cant know the ultimate
outcome in advance.
While most of us dont understand how to work with genetic
code, we have already begun to understand how to work with a
different type of code: software code. The design and engineering of software has changed quite radically in the last 10 years to
be much more open-ended than it used to be. In a sense, this is
a metaphor for how we as designers whether we are industrial
designers or designers of businesses need to behave and work
going forward.

Tim Brown is CEO and president of IDEO and the author of

Change By Design: How Design Thinking Transforms Organizations and Inspires Innovation (HarperBusiness 2009). He blogs at
designthinking.ideo.com

Rotman Magazine Spring 2012 / 21