Just as we are impressed when someone with expertise seems to know just what to do in a difficult situation, we are also impressed when someone invents a new procedure on the spot. This chapter explores the way people use leverage points—a small difference that makes a large difference, a small change that can turn a situation around—to create a new course of action, notice something that may cause a difficulty before there are any obvious signs of trouble, and figure out what is causing a difficulty.1 If you have the task of moving a boulder, you can put your shoulder against the stone and strain to overcome its massive inertia, or you can survey the scene to find the opening that might let you exert a small amount of pressure to shift the center of gravity and topple the boulder. The search for leverage points is about finding those openings and making them work. We also want to find our vulnerabilities, the easy ways our plans can collapse, so that we can take steps to prevent the difficulty.
Norman Berlinger (1996), a physician, describes a case in which a fetus was diagnosed as having a large cystic hygroma, a tangled mass of lymph vessels, on the side of his neck. The sonogram suggested that the hygroma had grown inside the neck, wrapping around the trachea. If that were indeed the case, without emergency treatment the infant would die shortly after delivery because his air passage was blocked. As long as the fetus was in the uterus, receiving oxygen from the umbilical cord, he was safe. A cesarean section was scheduled for the following day.
During the delivery, Berlinger was going to determine if the baby was able to breathe on his own. If he could not, Berlinger’s strategy was to intubate the infant, guiding a slender tube down his throat to clear a breathing passage. But that might be impossible if the hygroma had grown so large as to obscure the opening of the windpipe. In that case, it might be necessary to do a tracheostomy—piercing the trachea and inserting the breathing tube into it. That would also be difficult. A tracheostomy is easy with an adult, but an infant’s trachea is less than a quarter-inch wide, and soft and spongy. It is difficult to find. And other complications were likely. The incision would probably cut into the hygroma, possibly leading to infection of the lymphatic cysts and serious problems with abscesses in his chest. Moreover, the tracheostomy would have to be done under crisis conditions. It would be a last resort.
Upon delivery the infant gave a cry, suggesting a clear breathing passage. But then the passage sealed up. The infant could not even grunt. One of the nurses suctioned the infant’s mouth and nose and placed him in front of Berlinger. Berlinger remembered an earlier situation, when he had been called in to operate on a young man who had run his snowmobile into a strand of barbed wire strung above the ground to discourage trespassers. The wire had jumbled the victim’s neck tissue into sausage-like chunks. On that occasion, when Berlinger arrived, he found that the emergency technician had already inserted a breathing tube, and Berlinger had wondered how this was done. The technician later explained that he stuck the tube where he saw bubbles. Bubbles meant air coming out.
So in the delivery room, Berlinger looked into the mouth of the infant for bubbles. All he saw was a mass of yellow cysts, completely obscuring the air passage. No bubbles. Berlinger placed his palm on the infant’s chest and pressed down, to force the last bit of air out of the infant’s lungs. Berlinger saw a few tiny bubbles of saliva between some of the cysts and maneuvered the tube into that area. The laryngoscope has a miniature light on its tip, and Berlinger was able to guide it past the vocal cords, into the trachea. The infant quickly changed color from blue to a reassuring pink. The procedure had worked.
This example involves high-pressure problem solving. The physician did not have any procedures for inserting the breathing tube into the infant. He did recall an analogous case—a far-fetched one at that. It was about an adult, not an infant. It was about an accident, not a birth. It was about someone else’s actions, not his own. The key point of similarity was discovering the air passage in an obscured throat. And even the analogy was not sufficient. There were no bubbles. The physician had to invent his own way to produce bubbles.
Skillful problem solving is impresive because after the fact, the solution seems obvious, yet we know that, without any guidance, most people would miss the answer. They would not even know that an answer was possible. We would look into the infant’s mouth, see the mass of cysts, and abandon the idea of inserting a tube, immediately turning to the risky tracheostomy procedure, with no time to spare.
A leverage point is a focus for building a solution. It is the starting point for insightful problem solving. In example 8.1, the leverage point for the physician was the idea that bubbles could be used to find a breathing passage that otherwise could not be seen. In example 5.1, the leader of the emergency rescue team improvised a new way to rescue the motorist who was trapped in his car after it crashed into a concrete support. Some goals stayed the same (make the rescue quickly, do not endanger the victim, allow the victim to receive medical treatment), but some shifted. The goal of using the Jaws of Life to pry open the car doors was replaced by the goal of lifting the victim safely through the open top of the car. The strategy emerged on the spot, as both a goal (lift the victim through the top) and an action sequence (remove the top of the car, support the victim’s neck and back, lift and turn the victim, and so on). The leverage point was the realization that the roof posts were fairly well severed, so that the rescue could be made from above the victim. In that study of decision making, we coded the incident described in example 5.1 as an example of constructive problem solving, not as a recognition-primed decision.
One way to think about leverage points is to use an example from the sport of rock climbing.
To move from one place to another, a climber makes use of holds. What are the features of a hold? It turns out that holds do not have context-free features, such as protruding at least two inches and extending at least four inches across. A hold, rather, is whatever I can use to propel myself forward. If I can kick off with some power and do not have to travel too far, the hold can be smaller. On a more horizontal surface, at the start of the climb, a simple ridge less than a quarter-inch wide may do. If my fingers are tired, the protrusion needs to be larger. If it has been raining, or the climb is more vertical, or I do not think the rock is strong enough, I need a larger surface. A hold doesn’t have to protrude at all; a crack into which I can insert my hand will work. If the crack is too narrow to accommodate my hand but I can insert a finger and bend it so that the knuckle can be used as a brace, that may work too.
For these reasons, no one can examine a photograph and identify the holds. What counts as a hold depends on the conditions of the climb and the climber. Similarly, in problem solving we build our plans and solutions from the holds that we recognize.2
My colleague Steve Wolf and I first noticed the use of leverage points in a study of chess players. Sometimes the chess players were trying to reach a goal (“If I can only get that rook off that open file”), and other times they would notice a happy accident. For example, one player was mentally simulating a sequence of moves and noticed that he could attack his opponent’s queen. He did not find any way to capture the queen but started thinking of how to use this attack to gain some other advantage. The leverage point was in realizing that there might be an advantage gained in pressing the attack on the queen.
The decision maker, now acting as a problem solver, can try to use experience to identify leverage points and construct a new course of action. In evaluating whether the action will work, the decision maker now can try to improve it, or try to work with a different leverage point. Another approach is to try to modify the goals, which may result in the identification of different leverage points. The evaluation process can alter the understanding of the situation itself.3
Leverage points play an important role in a wide variety of domains. There are many examples from the field of business.4 I present just a few illustrations, starting with the idea for pocket radios. In 1952, Masaru Ibuka, working for Tokyo Tsushin Kogyo (later to become Sony Corporation), saw an opportunity to use transistors for consumer products. He believed his company could somehow build and market a transistor radio that could fit in a shirt pocket. In the early 1950s, radios depended on vacuum tubes; transistors were largely restricted to military projects. Most people thought the idea of a transistor radio was impractical. We now know that the concept of using a transistor for a consumer product was a leverage point. Developing, applying, and building on that insight had tremendous implications.
The realization by Boeing engineers that commercial jetliners would have a big advantage over propeller-driven planes was another use of leverage points. The engineers had not designed such a plane and did not have a ready market for one. They had a concept and the interest and curiosity to see what would happen if they were to put jet engines into airliners. Rival companies such as Douglas Corporation (now incorporated into McDonnell Douglas) were not motivated to build on this leverage point to design their own jetliner, and they were not prepared to handle the market that opened up after Boeing introduced the 707.
Henry Ford’s realization in 1907 that he could manufacture automobiles using mass production to cut costs so that most workers could afford one is a well-known example. At the time, Ford was just one of thirty competing automobile companies. His suspicion that mass production could cut costs dramatically was the leverage point.
In the early 1960s, Thomas J. Watson, Sr., at IBM, realized that a state-of-the-art system (to become the IBM 360) could change business in the way Henry Ford’s standard model automobiles changed transportation. IBM invested more money to build the 360 model than the cost of the Manhattan project to build the first atomic weapon during World War II.
Peter Schwartz (1991) describes a project he performed for Royal Dutch/Shell. The question was about the Soviet Union’s future policies on selling petroleum. Schwartz wondered whether there was any reason for the USSR to make dramatic policy changes. In reviewing some demographic data, he realized that the USSR could be headed for a sudden crisis. The population of elderly was increasing sharply, but the population of young adults entering the working class was decreasing. Schwartz wondered what this discontinuity might mean, and in exploring the implications, he was able to forecast, years ahead of the event, the possible destabilization of the USSR. He had discovered a leverage point for predicting a society under strain.
Scientists also rely heavily on leverage points. In a study of how successful human factors scientists and engineers first came up with their ideas (Klein & Hutton, 1995), we found that they were able to identify leverage points as opportunities where a need existed and the technology had become sufficiently mature. In other words, the scientists were not single-mindedly pursuing important topics. Unless they had a sense of how the problem could be solved, they did not get engaged. The leverage point provided them with a sense of the solvability of the problem. For example, for miniature visual displays, color would be better than black and white. However, color displays have lower resolution than black and white because it takes a mixture of three primary colors to produce a single pixel. Dave Post, a specialist in visual perception, wondered if he could use a subtractive method rather than an additive method, to get the same resolution as black and white. This hunch was enough to keep him working on the project and scavenging for funds for several years, until the technology was worked out.
In another example, Lee Task, a specialist in optics, noticed that pilots using night vision goggles could not read their instrument panels while landing their aircraft. He believed that the key data elements on the instrument panel could be incorporated into the night vision goggle display, so he identified this to his sponsors as a work project. Within a few months, he had developed a successful prototype.
Military commanders also need to detect leverage points. They need to find ways to exploit enemy weaknesses and detect signs that an adversary is preparing to do the same to them. The concept of maneuver warfare is to discover efficient opportunities for achieving powerful results. (It stands in contrast to attrition warfare, the approach of using force-on-force engagements to wear the enemy down.) A skilled commander is able to study a map and quickly detect the leverage points. Clausewitz refered to this capacity as the coup d'oiel, the rapid size-up that identifies the critical points in the terrain.
Leverage points are just possibilities—pressure points that might lead to something useful, or might go nowhere. Expertise may be valuable in noticing these leverage points. Certainly in games like chess, the experts are more likely to see them. In interpreting situations, experts seem attuned to the leverage points—both the opportunities and the threats facing them—rather than just being aware of the physical and spatial arrangement of objects.
In 1996, the world chess champion, Garry Kasparov, defeated the IBM chess computer Deep Blue in a six-game match. Observers noted that Deep Blue never adjusted its playing style. It always searched for the best move, even in positions where it knew it was behind. A human would have gambled on a strategy that was speculative rather than marching off to defeat. One of the IBM team members explained that the computer did not have a sense of “creative desperation”—the sense that drives chess players to search for leverage points, no matter how risky. In 1997, Deep Blue came back to defeat Kasparov. The machine still had not acquired a sense of creative desperation, but its developers had more than doubled the number of moves it searched and added some other refinements to make up for its limitations. The following example shows the use of creative desperation in spotting a leverage point.5
Israel is being invaded by one of its neighbors. The Israeli Defense Force commanders decide that they need to infiltrate tanks across the battle lines to conduct a critical maneuver, but they cannot find an acceptable place for the tanks to cross. One of the commanders scrutinizes the map and points to the place where the tanks will cross. “Here,” he says, “where it is marked ‘Impossible.’ That's where we cross.” He knows that the terrain has been tested with platoons, which consist of four tanks, and they have failed to make the crossing. Accordingly, the map was annotated to show the failure. But he also knows the terrain sufficiently well to realize that they can send one or two tanks at a time and make it through. And they do.
This theme is repeated in many domains. Experts know how the official records are compiled, whether they are maps, computer manuals, diagnostic tests, or air crew checklists. They know when the steps have to be followed and when to make exceptions.
Leverage points provide fragmentary action sequences, kernel ideas, and procedures for formulating a solution. Experts seem to have a larger stock of procedures that they can think of to use as starting points in building a new plan or strategy. These can serve as holds to get the process moving, or in combination with other fragmentary actions. Novices, in contrast, are often at a loss about where to begin.
We also need to spot leverage points that can work against us, in order to learn the weaknesses in our plans. These are sometimes called choke points. For example, a manager of a production line might realize that the whole schedule could be disrupted if a key part is not delivered on time. At that moment, no problem exists, but rumors about an impending strike at the plant that is producing the part cannot be shrugged off. It may be prudent to find a backup supplier, just in case. By noticing leverage points that can work against us, we buy ourselves time to take preventive action before the emergency arises. Example 8.4 contrasts two pilots: one who spotted a vulnerability and one who did not.
An airline crew takes off for a trip across the United States, from San Francisco to Philadelphia. A trained observer (himself a pilot) is riding in the cockpit, in the jumpseat. Before the airplane takes off, the crew conducts an inspection. The observer notes that the brakes and tires are fairly worn, almost to the limits for replacement. Once the crew members take their places in the cockpit, the preflight information package shows that they are carrying an extra 9,000 pounds of fuel. The observer asks the captain if he knows the reason, but it is a mystery. A few minutes later the captain radios the question to the dispatchers, but they do not know either.
After takeoff, about midway through the flight, the crew sees some flickering of the thrust reverser light. They do not believe that the thrust reversers are going to be accidently deployed, but the possibility exists that the thrust reversers may not be fully functioning to slow the aircraft down during the landing phase. Subsequently the crew learns that the weather in Philadelphia has deteriorated, with moderate winds and rain. The weather is not expected to get any worse, so the airplane should not need to divert. As they approach Philadelphia, the air traffic controller tells the crew which runway to expect for landing, so it can plan its approach.
The observer, realizing that the air traffic controller has selected the shortest runway at the airport, becomes disturbed: their stopping distance margin for error has been compromised, first by the worn tires and brakes, then by the extra weight, then by the possible thrust reverser malfunction, then by the wet runways, and now by the shortness of the assigned runway. Although still within the design limits for operating the airplane, the chances of stopping within the confines of the assigned runway have been reduced beyond his level of comfort. Knowing that the surface winds allow a longer runway to be used, the observer expects the captain to request it. Minutes go by, however, and the captain appears to be accepting the assigned runway. Wishing to call attention to the situation and to indicate his growing concern, the observer asks, “Which runway is that? It is that real short one?” The captain says that it is. Shortly, the captain initiates the radio call to air traffic control requesting a change in runway, which is granted.
The example shows two reactions to a potential problem. Both the captain and the observer had access to the same information. Both realized all the ways in which the landing was becoming risky. The observer identified the situation as problematic, requiring nonroutine action. In contrast, the captain appeared to be ignoring the warning signs and preparing to follow the air traffic controller’s instructions.
Let us go back to the example of the rock climb. We have used the holds to make it up the initial part of the climb. Now the route is becoming more difficult. We look up to where the old path continues and wonder if we can find a series of holds to make it that far. There are some possibilities, but they look risky. Then we notice that a crack has opened up that would let us cross an area that used to be impassable and bring us to a section that always seemed like an interesting challenge, if we could only reach it. This seems to be the time to try it out. We cross over using the new crack that opened up and now have to find a way forward. We scan the rock to discover the possibilities it offers. Each time we notice a good hold, we use it to map out how the sequence might go.
Once we come up with leverage points, we need to fill in the remaining details. In the harness rescue example, the commander synthesized a course of action by starting with the fragment of attaching a ladder belt to the semiconscious woman. From that starting point, he worked out the details of how to attach the ladder belt to her and how to tie a rope to the ladder belt to lift her to safety. By adding these details, the commander had the new course of action.
In the rock climbing example, we need to tie the leverage points together into a path in which we can feel confidence. Once we see how to move from hold to hold, we have a plan. We also know we will be changing the plan during the climb as we notice new features. We may have gaps in the plan where we do not see the connection but trust that we will find it when we get that far.