four such questions solves the problem. If you suddenly
start vomiting, you might guess that it has something
to do with what you ate last night—and you might be
right. But you might not.
Solution-guessing is a hit-or-miss technique. When a
problem has two or three potential root causes, and when
testing them is cheap and quick, it’s entirely appropriate.
But these are easy problems. Most persistent problems in
our lives aren’t easy by definition: They would not persist if they were easy to x.
What would we do if the breaker wasn’t flipped? Or
if it flipped again after a few minutes, plunging us once
more into darkness? Or if our light bulb blows out repeatedly? At this point, it’s time to realize we don’t have
an easy problem on our hands, and guessing won’t solve
it. If you don’t have a strong problem-solving skill set,
you have three options: You might keep guessing, hoping
you might resolve it. You might call in an expert—in this
case an electrician—and they’ll be able to use their experience to make an “educated guess,” which can move
easy problems along. But when that fails, you’ll probably
just cough up the money to replace whatever appears to
not be working, or just live with it.
When you’re facing a problem of moderate difficulty,
there may be something like 50 potential root causes. Perhaps you’ve developed intermittent sneezing fits, or your
motorcycle engine occasionally stalls out in the middle
of the highway, or you’re not making any progress on
your diet. At work, perhaps your emissions are too close
to the regulatory limit for comfort, or you suspect your
sales force is not selling as hard as they can because they
believe the supply chain won’t be able to meet their commitments to the customer. If you are really good at guessing—perhaps with the help of some colleagues—you
might come up with 30 potential causes.
It takes time and resources to test every guess. With
a long list, it’s likely you’ll waste lots of both. Worse,
there’s a good chance that the root cause isn’t on your
list, and you have no way of knowing until you’ve
completed testing the entire thing, which might take
months. What will you do next? Perhaps get a bigger
group together to create a longer list of guesses?
Then you’ve got hard problems. These are the
kinds of problems that might have hundreds or
thousands of potential causes. The actual root cause
is obscure or hidden. Shearing pipes in your water
pipes might be due to invasive corrosive bacteria
introduced at the local river. Your trouble sleeping
might be caused by an allergy to yellow-6 dye in
your macaroni. You are unlikely to be able to guess
the causes to these, and trying to guess wastes a lot
of time. Trying to implement some of these guess-
es is a shot in the dark and quickly uses up huge
amounts of resources. Your brainstorming efforts
will generate a list of some dozens of “possible root
causes.” You’ll tirelessly grind through them and,
months later, have nothing to show for it. Worse
yet, with all of the random changes you have made,
you’ve probably created new problems.
Brainstorming might be useful in situations where
creativity is required. However, solving hard problems
is not one of these. Rather than having one person
guess at something, brainstorming is gathering a lot of
people together to group-guess, which adds the further
complication of groupthink and politics. Often this
guessing is covered up with an elaborate “process” for
prioritizing the guesses. You can do better than this.
At one food processing plant, they were making a
product in a plastic cup with a seal on top—the sort you
tear off in order to eat. Customers were getting moldy
food because the seals weren’t working properly. You
can imagine this was a fairly important problem for
brand and food safety reasons. This corporation had
invested heavily in Lean and Six Sigma techniques
and had a sizeable organization dedicated to solving
this problem. When we arrived, they had used a Fish-
bone-Diagram approach to identify over 200 potential
causes and ideas to x them (this was clearly a pretty
hard problem). On the surface, they had taken a very
structured approach, but in reality, it’s what I call
“structured guessing.” Any time you “come up with”
many things to check that could be the cause, you are
guessing (see Table below).
If you get from someone a list of 10 “potential”
root causes, they don’t know what’s happening. If
you’ve come up with 200, you have no idea at all
what’s happening. This number of ideas is far too
many to search through with any reasonable effort:
An individual or team is going to run out of time,
resources, and energy long before they get through
the list. And worse, when a team doesn’t understand
a problem or the system behind it, odds are good that
