in the business world, we know more than we’ve ever known.
We have scores of business consultants at our disposal,
numerous business books published each year, a burgeoning
training industry gobbling up more than $60 billion per
year in the U.S. alone, and a record numbers of MBA graduates
hitting the market each spring. Yet, with all this knowledge
at our fingertips, it’s safe to assume that each of us could
list multiple examples where our companies, and we, have
had a hard time getting anything done. It can be very difficult
to turn what we know into action.
Zine was privileged to sit down with Stanford professor
Robert (Bob) Sutton, and discuss the problem. How can we
be well trained and well informed, yet ineffective? Sutton
and coauthor Jeffrey Pfeffer ask just those questions in
their highly successful book, The
Knowing-Doing Gap: How Smart Companies Turn Knowledge
Into Action (Harvard Business School Press, 2000). In
his fast moving and witty manner, Sutton took us on a lively
tour through the landscape of organizational behavior, focusing
on the links (and gaps) between managerial knowledge
and organizational action, and innovation. He
offered his perspective on why, despite all this knowing,
there’s very little doing, and how elearning might help
fill—and in some cases widen—the
is no knowledge advantage unless your company also
has an action advantage!
If you know by doing, there is no gap between what you know
and what you do.
Failure is essentialthere is no doing
without mistakes, setbacks, and dead ends.
Zine: We have spoken with many people about elearning, people
with wide and varied levels of experience and insight. Many of
them repeatedly return to the topics of experience, action, and
doingthe major themes of your new book. If
you would, please give us your perspective on learning and doing.
my perspective, the hazard elearning presents has been around
for a long time: when talking—or listening—becomes a substitute
for action. This is like the old transfer-training problem, where
talking about it is different than when the knowledge is actually
going to be used. Talk is not enough, so sitting in the classroom
or learning something just by looking at it on the computer is
not enough either.
the extent that you need to learn something in the box, elearning
might actually make a lot of sense. But, to the extent that you’re
trying to learn something in the box from elearning and then apply
it in a different or a more complex way, such as anything from
biochemistry to learning how to lead a group of people face to
face, or give speeches, it doesn’t work very well. Again, just
like the old transfer-training problem.
Zine: You mentioned learning things in the box, and we will
call that the known. Would elearning work only when things are
in the box, when things are known, or would a problem occur when
things are both known and unknown?
think the problem occurs in both cases. A lot of it has to do
with context. On the one hand, you have formal, structured learning
delivered on-line, typically presentation slides with a little
window in the corner of the screen where somebody talks. For the
more formal elearning, the computer can sometimes be a problem
because you lose the interactivity.
the informal side, Ted Anton has written a very interesting book,
Science. He interviewed seven famous scientists about their
breakthroughs, and shows that because of the Internet and the
new ability to exchange a lot of information and collaborate with
people over the world, discoveries happen much more rapidly and
dramatically than before. For example, the discovery of new planets
and the Human Genome project, among others, are much facilitated
by powerful computers and the Internet. The Information Age is
enhancing innovation, especially innovation that comes from new
combinations of disparate fields. This is the more informal side
of elearning. In some ways, it’s a structured versus an unstructured
Zine: To take a slightly different track, we would like to
touch on known problems and solutions versus unknown problems
and solutions. It’s probably not the same as structured versus
unstructured because the question is, is this a problem that someone
has solved before or is this a problem that no one has solved
before? We may not see new inventions happening all the time,
but we do frequently see people tackling things that haven’t been
the telephone and mail have been around for a long time, I believe
information technology makes a lot of things happen faster. But,
the question is how differently does it make them happen? I don’t
think it makes them happen that differently. It’s certainly not
an automatic fix to have information technology come and replace
A theory in the organizational arena called institutional theory
argues that companies do things, especially training, to help
satisfy external constituencies, and not necessarily to make things
better. For example, Stanford requires that I take a class on
sexual harassment over the computer next year. This is required
for legal reasons; then Stanford can say they’ve provided it to
everyone. To be blunt, I think the course has very little effect
on people’s behavior. Another example is some research I did several
years ago at a large local fire department that had terrible racial
problems. The judge mandated a lot of training, and this was training
where they sat and listened to people talk about the importance
of increasing racial harmony—then they would go back to the fire
house and there would be no change in their behavior at all. In
those cases, it was just perfunctory training that the institution
had to do in order to satisfy some external evaluation.
Nobody likes to say this about training, but there are certain
kinds of training that actually help people do their jobs better,
and there’s another kind that does not affect what people actually
do, but only affects the company’s legitimacy. (This is, of course,
not in the perfect world, but in the world, as it exists today).
In the latter case, if a course is as badly taught over the computer
as it is face-to-face, why not do the one that’s cheaper? I think
there is both an empirical and a theoretical justification for
cheaper in the case of the latter.
Zine: eLearning, as you describe it, is still a very passive
exercise today. Could you speak about your personal experience
with distance learning at Stanford, and how it has compared with
the traditional learning experience?
and many other institutions, has a lot of pressure to do what
they call elearning or distance education. There are all sorts
of offers for technology alleging it will help people learn better,
or at least to increase the profit margins of these institutions.
Those two goals get confounded sometimes.
For some sorts of activities, like just delivering a lecture where
the audience is completely passive, not much changes with the
on-line delivery tools of slides and a talking head. Of course,
it might be slightly better to work live with people around you,
and have a sense of the room, but not much is lost.
The more difficult arena is that middle ground where you’re trying
to get people to interact when some of them are physically with
you and some are on the network. People tell me that in some programs,
they entertain on-line questions and use phone calls, but in my
experience it has always failed, or been a less than robust experience
with lag problems and the like.
For example, in a typical class I teach 60 people live in class,
and 50 or so people via the web. Inevitably, tensions arise between
the in-class people and the out-of-class people, because communication
lags occur for the online participants and they usually miss subtleties
of the class. Of course, that’s only one way that distance education
Another method is that in a number of classes I teach now, I am
videotaped, and then the class is shown on-line here at Stanford.
Personally, that method is easier because I don’t have to worry
about a live external audience as I am being taped.
Zine: Have you seen elearning effectively applied, uniting
the knowing and the doing?
talks about all the fancy software applications for elearning,
but the most effective I have seen is straight email. For example,
I teach a class (with Tom Meyers) to a special group called the
Mayfield Fellows, 12 students with an entrepreneurship focus.
This course truly does a great job of marrying the knowing and
the doing, and it is much better because of technology.
Zine: Could you give us some real-life examples of how
the blending of the knowing and the doing occurs?
To illustrate, for the Mayfield Fellows, we teach live classes,
and then follow the students over the summer while they work in
a start-up. They keep in constant touch with each other and with
us by sending email and writing in diaries once a week about what
they’re doing. Reading the diaries and the emails during the summer
gives all of us a much richer feel for their experiences, and
we’re active participants by offering feedback, advice, and our
own experiences. So, in this case, people are actually doing something,
but in twelve disparate physical environments.
Zine: So the most effective application you’ve seen is really
guess my bottom line on elearning, and related products, is that
it shouldn’t be a substitute for actually thinking and doing something.
However, when interwoven with what people actually do, it can
make the learning experience richer. It is not a replacement,
but rather a supplement, and it can make learning more efficient
and sometimes richer.
As I think about it, maybe that’s the distinction. It will not
work as a substitute for actually doing anything else—the
old problems persist. You can’t learn heart surgery or how to
fly an airplane on the computer screen, nor learn by watching
somebody do it. It will work if it somehow increases the flow
of communication and understanding among people actually doing
something, just like our twelve students struggling to be members
of start-ups this summer. They can read each other’s diaries,
and they can send emails to one another and ask for help.
Zine: If you were going to help educate the world on some
of the messages of the Knowing-Doing gap, and help people and
companies get started turning knowledge into action, could you
offer some of the first action steps, or things they can do to
can point to two of the most serious problems that make it difficult
for organizations to actually turn knowledge into action. One
is called the smart talk trap, or when talk becomes a substitute
for action. It seems to me that the hazard of on-line education
is that we will simply continue that problem. I don’t think on-line
education will make it worse. But when you believe that you’ve
trained somebody to do something just by talking about it rather
than learning by doing it, you’ve made a mistake. As I mentioned
before, that’s the same mistake that training programs have always
The second cause of the knowing-doing gap is precedent, which
is when people do things in organizations because they’ve always
been done that way. One thing I worry about with some of the on-line
education is that it makes the company’s dogma even clearer and
more explicit. There’s some risk that people in companies will
become more efficiently frozen in a path and won’t be able to
try new things very easily. That is, of course, a problem in organizations
anyway. I worry that by codifying everything, writing everything
down, and delivering it on-line that it’s harder to have people
make the same mistake again and actually discover that while it
didn’t apply in 1997, it applies now. So, we’re ironically talking
about increased efficiency getting people stuck in a certain path.
Zine: We could have a whole conversation about un-learning,
un-doing, and learning from our errors...
of these advances in technology might make it harder for people
to un-learn, and to do random things. In many cases, random things
are actually quite important to innovation. So again, one of the
things I worry about is how elearning will more efficiently present
uniformities to people.
With that said, people are inquisitive and very curious. I think
they can overcome virtually any machine and figure out a new way.
Zine: Well it’s been a pleasure talking with you. We are anxious
to begin doing!
Sutton is a Professor of Management Science and Engineering in
the Stanford Engineering School, where he is Co-Director of the
Center for Work, Technology, and Organization and an active researcher
in the Stanford Technology Ventures Program. He is also a Fellow
at IDEO Product Development. He has served as an editor and on
the editorial board member of numerous scholarly publications,
and currently serves as Co-Editor of Research in Organizational
Garlington Scofield is managing editor of LiNE Zine. She welcomes
comments, challenges, and questions about anything she's said.
Write her at email@example.com.
Copyright (c) 2000-2004 LiNE Zine (www.linezine.com)