|
This
article originally appeared in the March 1998 issue of
PDBP
EXPERT INTERVIEW with Preston
Smith:
Faster, Better, Cheaper Not Always The Best
Policy
In 1991, when Preston Smith and Donald Reinertsen
published their best-selling book, Developing Products in
Half the Time, thinking about cycle-time reduction as a
key strategic lever was just starting to hit center stage.
That was the period Tom Peters dubbed the "Nanosecond
Nineties" in his then-popular video, Speed Is Life: Get
Fast Or Go Broke. Times have changed, says Smith. And in
the process too many executives responsible for product
development decisions have bought into a misguided notion that
cycle-time reduction is, by itself, a universal panacea. As
the authors prepare to launch a new edition of their book
(published by Van Nostrand Reinhold and available in
bookstores later this month), we caught up with Smith to ask
him about what’s changed:
BPR: Despite the
title of your book, you raise some cautions about the
relentless, unthinking pursuit of better, quicker, cheaper.
What are those cautions?
PS: "When the
original edition appeared in 1991, few managers were thinking
of rapid development as a competitive weapon. Now, most
managers seem to believe that time compression is always to be
pursued, along with quality and low cost. But time to market
is not a universal solution to product success. Instead, it is
a set of tools that have both their value and their cost.
Every manager and developer should be aware of the tools but
apply them only when their benefits exceed their costs.
The final measure is profit, not speed. For each
development project, the blend of speed, quality, and cost
will be different. Unbridled pursuit of speed and product
proliferation will drive you into the poorhouse, as we have
seen with much of the Japanese consumer electronics
industry."
BPR: Who needs to
read your book these days?
PS: "Anyone
who is involved with the development of new products. This is
not an engineering, a marketing, or a manufacturing book, but
it involves elements of all three. Most companies will find
that the biggest opportunities for improvement lie at the
interfaces of these functions. Because we have seen the best
adoption of these techniques in companies that train everyone
who touches a new product, we have arranged with the publisher
to offer the book at attractive discounts to companies that
buy it in quantity.
"I have great
difficulty listing in a few words the industries to which the
book applies, because it fits most industries, from mature
ones, such as food packaging and consumer durables, to high
tech ones, including aerospace, computers, and medical
devices. The only limitation is that there should be some
science or engineering in the product itself, although
insurance companies have used it for their ‘products.’
"The real question is
not whether the industry or market segment is a fast or a slow
one, but rather how quick your company is relative to its
competitors in meeting customer needs. Chrysler is beating its
competitors in a very established, high-inertia industry by
using time to market effectively."
BPR: What is
strikingly new in the new edition?
PS: "Nothing
is strikingly new; the original formula has proved to be
remarkably durable. The ‘New Rules’ that now appears in the
book’s title refers not to new rules in the book, but instead
to new rules in the marketplace that have made time
compression a more valuable and sophisticated competitive
weapon.
"Although the
skeleton remains the same, the treatment of each of the
fifteen chapters has matured and broadened considerably, and
there are plenty of new tools. For example, our experience
since the first edition has convinced us that the economic
tools (cost of delay) in Chapter 2 are the very foundation of
rapid product development, so these tools are more thoroughly
integrated into the other chapters. And Chapter 2 is now much
more useful, with information on keeping the economic model
simple, and how to expand it in certain circumstances, such as
long-life products and interacting products. The chapter also
shows how to do a ‘sanity’ check on the
model.
"Chapter 4
(incremental innovation) in the new edition gives equal
treatment to both the pros and cons of innovating
incrementally, so readers can apply this tool wisely. Chapter
5, on product specifications, is now far more customer-focused
and includes better guidance on quality function deployment.
Chapter 7 has valuable new information on motivating
development teams, Chapter 8 expands on dispersed work groups
and virtual co-location, Chapter 9 shows how to design fast
processes and integrate technology tools (for example,
computer-aided design) , and Chapter 12 provides a
comprehensive approach for managing project risk. Readers will
find about 50 percent new material in every chapter, while
maintaining the complete book at its original length, for ease
of use.
"Besides increased
emphasis on the cost of delay as the foundation, you will
notice stronger attention to synergy in combining the tools.
Product development has become more demanding since 1991, so
using a few isolated tools no longer provides sufficient
competitive advantage. We have also provided more material on
software development, which in many ways is quite similar to
hardware development, although many programmers would
emphasize more the differences with hardware development.
Because the book in many cases has become a dog-eared
reference for developers, both the table of contents and the
index are now more comprehensive, and icons in the margins
direct readers to key points."
BPR: You argue
for buying cycle time at the right price; how do I know the
right price?
PS: "You will
not get past Chapter 2 without knowing how to make and check
these calculations. And if you try to skip this chapter, you
will be nudged gently back to it."
BPR: Aren’t the
rules fundamentally different for different
sectors?
PS: "Yes and
no. Certainly, one’s approach to managing risk or surveying
customers is quite different when developing an airliner than
when doing a computer mouse. Hewlett-Packard’s ability to
co-locate a development team differs radically from that of a
start-up company. This is why we operate at the level of
illustrating the advantages and pitfalls of a variety of
tools, so each organization can select and modify tools to
meet its own needs.
"But at its roots,
this subject gets down to changes in operating styles and
human behavior and attitudes, which carry across all sectors
and countries. This is why we highlight areas that are likely
to require cultural changes, and we provide a whole chapter
(Chapter 15) on making these changes. Although we have written
the book modularly, so that the reader can easily pick and
choose, my own experience is that the topics that often seem
most insurmountable for a particular company are precisely the
ones that are likely to have the most potential. So I try not
to lose sight of these in a particular setting, even though it
is easy to dismiss them. These ‘growth opportunities’ are
unique to each company in its journey toward improved product
development."
BPR: You’re an
advocate of disciplined continuous learning from projects and
to improve the process. What’s the issue
here?
PS: "Precisely
because each company is at a different spot, the richest
sources for possible improvements in that company come from
its own experience. Just as a company will never become a
leader by copying its competitors’ product offerings, it can
never have a leading development process by adopting the ‘best
practices’ of others. A development process that arises from
continuous learning will be the most efficient approach for a
given company, it will be a competitive advantage that cannot
be taken by its competitors, and it will prepare the company
to deal better with changes in the competitive environment
through the process’ built-in
flexibility."
BPR: Doesn’t this
call for longer-term commitments than most organizations are
willing to make in the Nanosecond Nineties?
PS: "Yes.
However, companies that intend to be leaders in the next
century— such as Black & Decker, Chrysler,
Hewlett-Packard, and Toyota—are making these commitments
today. I can see a real difference within my client base of
the companies that make these commitments and those that
don’t; my client list will look different in a few
years!"
BPR: How do you
begin to implement a cost-effective, disciplined,
value-adding, and non-punitive approach to organizational
learning from product development
activities?
PS: "To have
any staying power, the program needs to have the leadership
and backing of upper management. It will have its cost, simply
because any effort put into reviewing and improving the
process will have to come out of effort that could be put
directly into developing products by the same people. This is
essentially an investment in the future, but an investment
that I see our better clients making. There is no doubt in my
mind that the investment has great long-term financial
payback, and management can decide how much they wish to
invest.
"To make the process
‘disciplined,’ and to help process review become an expected
part of doing business, it is important to review every
project (above a certain threshold in size). This will also
help with the ‘non-punitive’ aspect, but the key here is to
focus the reviews on the process, not the individuals
executing it. A good facilitator during the early reviews can
help keep the attention on the process. Companies such as
Intel and Microsoft have built a culture that allows people to
be quite critical of the process without devaluing the people
involved."
BPR: Metrics:
what are some common mistakes you see organizations
making?
PS: "Metrics
often get connected with benchmarking: companies try to
measure themselves against other companies. Product
development is so variable—even among projects within a
company—that trying to compare between companies is likely to
lead to so much noise that the comparison will have little
value. And, as I’ve already suggested, comparing oneself with
another’s past performance is bound to make one a
follower.
"I suggest looking at
two distinct classes of metrics. One is long-term trends that
allow a firm to measure its improvement by filtering out the
project-to-project variation. Hewlett-Packard’s vintage charts
(adapted in Figure 1-2 of the new edition of the book) are a
perfect example of a metric that HP has been using in a
constant format for over a decade to track progress toward a
corporate goal.
"The second class is
short-term metrics aimed at specific problem areas. In the
book, our HP example of such metrics involves how its
management monitors model shop queues so that it knows when it
needs to add capacity. The mistake often made here is to mix
these two classes; then, either definitions of the long-term
metrics shift before they can show trends, or the more
tactical metrics lose the flexibility they need for day-to-day
management corrections."
BPR: If there was
one single thing you’d urge product development leaders to
keep in mind about metrics, what would it
be?
PS:
"Start by being clear about where you want to go and why. As
that great corporate strategist, Alice in Wonderland, learned:
‘If you don’t know where you want to go, any metrics will take
you there.'" | 
