Page 129 - The Drucker Lectures

110 [ The Drucker Lectures

is the case, Stanley, please write out your resignation and nomi-
nate the brightest hamster. We’ll make him research director.’”
It took me six weeks to get across to Stanley what the presi-
dent had been trying to tell him. And he never quite accepted it.
A good many people still feel that way when you say “knowledge
work.” In other words, knowledge is not, in that sense, quanti-
tative. And so we will have to learn to think through how we
measure and how we appraise. And then I think we can begin to
focus knowledge work on results.
One result is productivity, which is woefully low—not be-
cause people don’t work hard, but because we don’t know what
productivity means. We made the same mistake with manual
work—measuring productivity by how much sweat there is,
how hard it is, how many hours are being worked, and how
unpleasant it is. Before [Frederick] Taylor, the main measure-
ment of productivity was how tired people were when they got
home. Well, that’s not the measurement of productivity; that’s
the measurement of incompetence. And we are doing that with
knowledge work.
Let me come back to the question I started out with—the
question posed by my friend, the research manager, over the
telephone a year and a half ago: “Can knowledge be managed?”
The answer is: We don’t know. But we do know that it has to
be managed, and if knowledge is not managed it only costs and
doesn’t produce. And we know that it has to be managed differ-
ently, that you must start out with a few uncommon assumptions,
counterintuitive because that’s not the way we look at other work.
You must start out with the assumption that knowledge changes
itself—that more you know, the more it changes. There’s the as-
sumption that, by itself, knowledge is an input, and it has to be
integrated to become an output. And there’s the assumption that
knowledge must be concentrated. If you splinter it, you get very
little. You get journalism, but not knowledge.