"SMART
PROTOTYPING"
Fast and Flexible?
Do It
Wrong the First Time
by Preston
Smith, New Product Dynamics
Computers allow us to do many
wonderful things today but, in many cases, our thinking
about how to use this tool wisely hasn’t caught up with
rapid advances being made in the technology. It’s like
the horseless carriages of a century ago: they still
looked like, and were operated much like, horse-drawn
carriages!
A similar thing could be said
about prototyping, a vital tool of product
development. Prototyping has received more attention
lately, in part from Professor Stefan Thomke of Harvard
Business School (See "Enlightened Experimentation: The
New Imperative for Innovation" in the February 2001
Harvard Business Review). He and others are
finding that the new prototyping tools available today
can only be exploited by rebuilding our product
development process – and our mindset – to take
advantage of what modern prototyping can
do.
For example, using modern
computer technology, you can "do it right the first
time" by completing a great deal of design, analysis,
simulation, and checking of a new part on the computer.
This means that the first time you actually make the
part, it is "right," at least in an engineering sense.
As the engineers are going down this route to
perfection, however, the design is hidden from
customers, partners in the distribution channel,
suppliers who will have to make it, and even the test
engineers who must design a test for it. Engineers – all
of us, actually – prefer this approach because it
comfortably shields us from criticism of our half-baked
ideas.
What Thomke, et al.,
are discovering is that it is actually faster, and more
accurate, to do it wrong the first time. Get
something – almost anything – out there and let others
react to it and tell you what is wrong with it before
you go down a path that won’t delight them. Expose your
ignorance, but do it quickly so that you can then move
forward more surely. Furthermore, this line of thinking
suggests that you keep exposing your ignorance as you
go, continually drawing in others whom you will
eventually have to satisfy.
Unfortunately, developers of
computerized prototyping tools are mostly pursuing other
objectives. Rather than rapid, cheap prototypes that
allow you to expose your ignorance early, they are
developing prototyping tools that will build an elegant,
expensive, "perfect" part later. There are some
important exceptions to this trend, but you will have to
search to find them, and you will have to be clear about
what you’re seeking.
What is more difficult is
that you and others in your organization will have to
think about product development and "failure" quite
differently in order to take advantage of modern
prototyping capabilities. It is useful to think of
product development as a series of decisions or forks in
the road. Prototypes and experiments help you navigate
these forks. The trick is to design the prototype or
experiment that gives you clear guidance at each fork so
that you can get past it, and on to the next one,
quickly and accurately. Whereas, in the past,
prototyping technology was generally too expensive to
make a prototype at every fork, today there are
affordable technologies that enable developers to do
just this. The figure on this page illustrates the
difference between making a few elegant, expensive
prototypes versus numerous expendable
ones.
Using this mode of operation,
each prototype is aimed at answering only one question,
and it is only good enough to answer this question. When
the question is answered, you toss the prototype and
move on the next one. To do this effectively, you will
also have to be sensitive to building consensus and
commitment as you go. If you have traveled a great
distance in your journey, and an executive who hasn’t
been involved wants to revisit a much earlier fork in
the road, you will have lost much of the advantage of
this approach to prototyping.
One warning. This "do it
wrong" approach is not the best method in all cases.
For example, if you are designing a memory chip, the
requirements can be laid out objectively at the outset,
without further customer involvement, while the cost of
making a mistake is very high. In such cases, it pays to
get it right the first time. However, most product
development projects will benefit from a more
interactive approach, and cheap, quick prototypes are
perfect for this.
This article originally appeared in the May
2003 issue of PDBPR
 To join our e-mail list and
receive future updates on "Fast & Flexible"
techniques, click
here.
|
ABOUT
PRESTON SMITH