Creative People Must Be Stopped (23 page)

Conscious Competence: Knowing What You Know

Understanding the type of problem you face and having the tools to solve it is among the less challenging of situations. We use existing methods to solve problems that are exactly the ones those methods were designed to solve. Models and analysis, two tools of engineering and science, work in this way and are powerful for quickly finding solutions. If you were an architect designing a normal house in a normal neighborhood, you could quickly calculate, using models developed in the profession, how thick the beams holding up the roof will need to be. Sometimes it won't even seem as though you are doing an analysis, especially if you are simply looking up the answer in a table based on predefined parameters.

This type of problem is also commonly addressed in business, where “management science” offers defined approaches to solving problems related to product pricing, market segmentation, financing, inventory management, and even employee motivation, among others. Once we have developed an understanding and are clear about the limitations of that understanding, one aspect of the constraint will be that of “doing the math.” It may be difficult to find the data you need to be able to apply the analytical model to your problem. Another facet of the constraint is that the solutions you will arrive at are those limited by the models. For example, a problem that airlines face is getting passengers into their seats as quickly as possible before takeoff, because—as the saying goes—the plane makes no money sitting on the ground. One approach to this problem could be to use a mathematical queuing theory model to determine the most efficient order for boarding people. However, this approach will never generate the solution Southwest Airlines used for years: it didn't assign seat numbers. Without assigned seats, people arrived early and then rushed on board to find the best ones. Coming up with such a nonobvious solution requires a wholly different kind of competence, one that you may not have access to.

Unconscious Competence: Not Knowing What You Know

For certain problems, we may intuitively know what needs to happen and why it needs to happen that way. In
Shop Class as Soulcraft
, Matthew Crawford (2010) shares his experience of moving from a job at a think tank to becoming trained as a mechanic of classic motorcycles, and his condition as a mechanic fits this type of technological situation. When someone brings in an old bike for repair, Crawford often doesn't know what's wrong with it, how much it will cost to fix it, or even whether he can repair or restore it. But he takes it in anyway because he is an expert mechanic and has worked on similar bikes before. To effect the repair, he goes through a sophisticated diagnostic process that involves calling on his prior experiences, rules of thumb, traditional understandings, unjustifiable assumptions, ad hoc experimentation, and dumb luck. And in the end, the insights and conclusions of these efforts will have to be actionable; that is, he has to know how he will do what he has decided needs to be done.

Apart from the need to rely on intuition and untested assumptions, a primary constraint inherent in this situation suggests a kind of closure of the circle of Maslow's model. Our success in an unconscious competence mode makes us susceptible to an imperceptible slide toward thinking that we know how to solve problems that we actually don't. This brings us back to the start, in the dangerous place of unconscious incompetence.

Overcoming Physical Constraints

If anything is true about the difficulty of pursuing technological innovation, it is that your chances of success are far lower if you rely on happenstance or luck than they will be if you make significant investments of money, time, and hard work. But as you go about making those investments, you may wish to consider the following strategies.

Determine How Hard the Problem Might Be

The most important strategy for overcoming technological constraints is to start by knowing what you know and what you don't know—that is, to understand how difficult a problem it is that you are trying to solve. With this understanding, you can try to make reasonable estimates of the time, money, and people resources you are likely to need.

Consider the problem of deciding how much money to spend on online versus print advertising. For this problem of
conscious competence
, a person trained in marketing science (or a reasonable facsimile thereof) might be able to analyze your existing expenditures and resulting outcomes and then come to a reasonable conclusion about what to do going forward. Intuitively this problem seems relatively straightforward, should require only small amounts of resources, and could be rapidly achieved. In contrast, a
conscious incompetence
problem like developing the atomic bomb on the basis of the newly developed and still evolving field of quantum physics is obviously going to be terrifically expensive and will take a long time. And the output won't be guaranteed.

To make a first cut, you can use expert opinions, prior experience, and even the patent history in a specific domain to get at least a starting estimate of how difficult the problem is likely to be, given our current state of understanding the world. Of course this is going to be a very rough estimate, but it may serve as a good start for estimating the kinds of time, money, and brains you will need to come up with to tackle the problem.

Next comes the hard work of determining the possible pitfalls associated with being wrong about your estimates of your levels of consciousness and competence with respect to the problem. How much of a setback will you suffer if you have overestimated your ability to implement the state-of-the-art solutions? How much harder or more expensive might it become if you have to develop an entirely new way of thinking if the current methods cannot get you there? And what is the risk that you have misspecified the problem or don't know something important about the problem? Of course this is a difficult line of thinking to engage and an even more difficult conversation to have with others, especially if you are deep in a process, running low on resources, and struggling desperately to keep the project going.

After doing a “reality check” assessment of the difficulty of the problem given your technical approach, the size of the team that you have at your disposal, and the budget you have of time and money, it may be time for a difficult decision. If the project is insurmountable and lies outside too many of the constraints of the type we've discussed in the preceding chapters and especially outside your organization's capabilities, then please stop. Remember, creative people must be stopped when they're going down the wrong path, one that leads them to tackling problems only because the goal is attractive to them and not because solving those problems furthers the strategic aims of the organization. Short of stopping, you might also reframe the problem and rethink your approach in a way that can still yield positive benefits, but that allows them to be achieved in a way that's possible given the people, time, and money that you have.

Hire an Alien

I once had an opportunity to tour the innovation center of a national health insurance company. As we walked around the facility, my guide pointed to two people and said, “Those two were working on the settee when we hired them.” I had to ask what “settee” meant, as I was sure he could not possibly mean what I thought I'd heard. He said, “You know, the S-E-T-I project.” I'm not sure which would have been more surprising to hear—that they were working on a couch or that an insurance company had hired experts in the search for extraterrestrial intelligence.

My guide explained that the company had realized that its role in the health care industry gave it unique access to immense amounts of data. As the people who handle the bills for a great many health care transactions, they were in a position to see if there were any interesting patterns or “intelligent signals” that might be found in the data.

They would not give me any further details—except to assure me that the data were stripped of all personally identifying information. Still, I could easily imagine learning some interesting things by examining the data. For one, you might look for patterns of treatments and then subsequent recurrences (or lack of recurrences) of the treatment for a particular ailment, giving you some ideas about the efficacy of a particular treatment. Or you might look at patterns of hospitalizations and rehospitalizations based on treatment types or on adherence to medication regimes. I'm sure that the patterns found in the data would have all kinds of potential meanings for people who were actually versed in health care.

What I do know about health care, though, is that analyzing multiple, massive data streams is not a core skill that physicians are trained in or that health care statisticians will commonly use. However, this kind of analysis is
exactly
what SETI researchers do all day and that they probably do in their sleep. My hat goes off to this company that recognizes the need for this kind of analysis, sees the value of bringing that skill in-house, and then finds the people who can perform it. After all, the organization is hiring people who think differently, analyze differently, communicate differently, and—for Pete's sake—look for intelligence in the universe beyond Earth.

Let Other People Do the Hard Part

Beyond the technological development that happens in schools, universities, and R&D labs, a great deal also happens among consumers who have a need to push performance boundaries by modifying and customizing the products they use. Eric von Hipple (1988) calls these people “lead users” and suggests that there is a lot you can learn from them.

Lead users may modify even the highest-end products, not hesitating to void the warranty in their quest for maximum performance. They might be amateur cyclists working to decrease the weight of a bicycle, boy racers fiddling with the software of the engine-control chips in their cars, or even purchasers of an industrial printing machine who, by tweaking the feed rollers just a touch, are able to get the machine to run a lot longer without jams. Some of them even end up starting their own businesses.

Many organizations frown on this kind of activity. And it is worth frowning on when performed by hacks or ignoramuses. But lead users are neither. They are simply people with a real need that a manufacturer has not learned (or bothered) to fill, and an intense motivation to fill it. Honor them by learning from them. They are doing the work anyway; you might as well benefit from it to speed up your own innovation processes. You might take an example from the companies that have legitimized or created online communities where individuals praise or malign their products in public. Users realize that these discussion groups are not controlled or censored by the companies and thus are a place to really tell businesses how you think. Many companies are scared to death of these boards and try to suppress them, building ill will and cutting off an important source of information. A more constructive alternative approach could be to make it the better part of
everybody's
job to monitor, take part in, and, most important, learn from their customers' discussions.

Invest (to a Point)

Technological innovation is, as I suggested at the outset of this chapter, an expensive proposition. R&D tends to require experts, equipment, time, and money, and often more of all of these than we imagine at the start. So if you're going to invest, do what it takes. As one company head said as he approved the biggest project budget he had ever signed off on, “If you're going to jump a chasm, you can't jump halfway.”

However, you also need to decide which chasms are worth crossing, or as Maslow (2000, p. 17) observed, “What isn't worth doing, isn't worth doing well.” I've already talked about the need to understand how difficult the problem is that you're addressing. If it is too hard (hence costly) relative to your means or the chances of success, don't jump. Maybe you can succeed, and maybe the innovation will genuinely improve things, but you may be at a point of diminishing returns. This is the point where additional improvements come only at the cost of substantially increased investments, and is possibly the point where a completely different technological approach is called for. If achieving the last 20 percent of improvement is likely to cost as much as, or more than, achieving the first 80 percent, then it is definitely time to ask whether it's really worth it to make it 20 percent better using the approach you are currently using. Of course, the answer depends in part on the size of the payoff: if it's big, the cost may well be justified. The key is to ask the question so that you are making a deliberate choice, not simply following the path you were on because you didn't think to question it.

Time Constraints: Having Time and Making Time

“Clock time,” that immutable force of nature that has to be dealt with when synchronizing and coordinating the actions of technological innovation, can be a significant constraint. We often treat it in this way, as if it were a completely external constraint, using a phrase like “There isn't time to try that,” which can translate as “End of discussion.” But our relationship to time is worth discussing because our assumptions about the amount of time we have, about when something needs to get done, or even about how we will use the time we have may create significant yet avoidable constraints that, unlike clock time, can be overcome.

Sequencing and Coordination Requirements

Certain complex projects or tasks require that things be done in a particular order. For example, you would not put the hubcaps on the wheels of a car before attaching the wheels to the axles, just as you would not hire a caterer without first deciding what kind of food you want to serve. There are other kinds of tasks for which the constituent parts can be performed in any order or even in parallel. When digging a ditch, I can start at either end, and additional help from someone starting at the other end can make the job go that much faster. Not recognizing the difference between these two kinds of tasks can be a significant source of constraint.