Chapter 4
The Not-Invented-Here Bias
Why “My” Ideas Are Better than “Yours”
From time to time, I present different research findings to groups of executives in the hope that they will use some of them to create better products. I also hope that they will, having deployed the ideas in their companies, share with me their results on how the ideas worked out.
During one such meeting, I offered a group of banking executives some thoughts about ways they might help consumers save money for the future, rather than encouraging them to spend their paychecks as soon as they get them. I described some of the difficulties we all have when thinking about the opportunity cost of money (“If I buy that new car today, what won’t I be able to do in the future?”). I proposed some ways in which the bankers might concretely represent the trade-off between spending now and saving for tomorrow to help their customers improve their financial decision making, and, in the process, increase their customer base and loyalty.
Unfortunately, the bankers didn’t seem terribly stirred by what I had to say. In the course of trying to interest them, I remembered an essay by Mark Twain called “Some National Stupidities.” In his essay, Twain praises the German stove and bemoans the fact that Americans continue to rely on monstrous woodstoves that practically require a dedicated full-time staff to keep them running:
The slowness of one section of the world about adopting the valuable ideas of another section of it is a curious thing and unaccountable. This form of stupidity is confined to no community, to no nation; it is universal. The fact is the human race is not only slow about borrowing valuable ideas—it sometimes persists in not borrowing them at all.
Take the German stove, for instance—the huge white porcelain monument that towers toward the ceiling in the corner of the room, solemn, unsympathetic, and suggestive of death and the grave—where can you find it outside of German countries? I am sure I have never seen it where German was not the language of the region. Yet it is by long odds the best stove and the most convenient and economical that has yet been invented.5
According to Twain, Americans turned up their noses at German stoves simply because they hadn’t come up with the better design themselves. Analogously, here I was, looking at a sea of unenthusiastic faces. I was presenting the bankers with a good idea—not just some vague notion but one supported by solid data. They sat back passively in their chairs, clearly not taking in the possibilities. I began to wonder if the lack of excitement on their part was due to the fact that the idea was mine rather than theirs. If that were the case, should I try to get the executives to think that the idea was their own or at least partially theirs? Would that make them more interested in trying it out?
The situation reminded me of a commercial that FedEx ran a while back. A group of shirt-and-tie-wearing employees are sitting around a boardroom table, and the more formally attired boss announces that their objective is to save the company money. A doleful-looking, curly-haired employee offers the following suggestion: “Well, we could get an online account with FedEx and save ten percent on all our shipping costs.” The other employees glance around in silence, waiting for a signal from their leader, who has been listening quietly, hands folded meditatively in front of his face. After a moment’s silence, he emphatically slices the air with his hands—and then repeats what his sad-eyed employee has just said. The other workers cheer sycophantically. The fellow who made the suggestion points out that he just said the same thing. “But you didn’t go like this,” his boss responds, repeating his emphatic slicing gesture.
To me, this humorous commercial demonstrated a crucial question of how people relate to their own and others’ ideas: how important is it for us to come up with an idea, or at least to feel that it is ours, in order to value it?
The attraction to one’s own ideas has not escaped the collective folk wisdom of the business world, and, like other important business processes, this one also has an unofficial term attached to it: the “Not-Invented-Here” (NIH) bias. The principle is basically this: “If I (or we) didn’t invent it, then it’s not worth much.”
Any Solution, as Long as It’s Mine
With our understanding of human attachment to self-made physical goods (see the previous chapter on the IKEA effect), Stephen Spiller (a doctoral student at Duke University), Racheli Barkan, and I decided to examine the process by which we become attached to ideas. Specifically, we wanted to test whether the process of creating an original idea is analogous to building one’s own toy chest.
We asked John Tierney, the science writer for The New York Times, to post a link on his blog6 asking his readers to take part in a study about ideas. A few thousand people followed the link, were asked about some general problems that the world is facing, and evaluated solutions to these problems. Some respondents proposed their own answers to these problems and then evaluated them, while others sized up solutions that Stephen, Racheli, and I came up with.
In our first experiment, we asked some participants to look at a list of three problems one at a time and generate their own proposed solution for each. (We called this the creation condition.) The problems were:
Question 1: How can communities reduce the amount of water they use without imposing tough restrictions?
Question 2: How can individuals help to promote our “gross national happiness”?
Question 3: What innovative change could be made to an alarm clock to make it more effective?
Once the participants finished generating their three solutions, we asked them to go back and rate each one on practicality and probability of success. We also asked them to tell us how much of their own time and money they would donate to promote their proposed solutions.
For the noncreation condition, we asked another group of participants to look at the same set of problems, but they didn’t get to suggest any solutions. Instead, we asked them to evaluate the solutions that Stephen, Racheli, and I came up with and evaluate our solutions in the same way that the participants in the creation condition evaluated their own solutions.
In all cases, the participants rated their own solutions as much more practical, as having a greater potential for success, and so on. They also said that they would invest more of their time and money into promoting their own ideas rather than any of the ones we came up with.
We were pleased to come away with this type of supportive evidence for our Not-Invented-Here bias, but we didn’t know exactly why our participants felt that way. For one thing, it was quite possible that their ideas really were better, objectively speaking, than the ones we came up with. But even if their ideas weren’t superior to ours overall, it could have been that our participants’ notions fit better with their own unique perspectives of the world. This principle is called an idiosyncratic fit. As an extreme example of this, imagine that a devoutly religious individual answered the question “How can individuals help to promote our ‘gross national happiness’?” by suggesting that everyone attend religious services daily. A steadfast atheist might respond to the same question by suggesting that everyone give up religion and focus instead on following the right kind of diet and exercise program. Each person may prefer his or her idea to ours—not because he or she came up with it but because it idiosyncratically fits with his or her underlying beliefs and preferences.
It was rather clear to us that the results of this first experiment demanded further probing. We did not know how much the increase in excitement over the participants’ own ideas was due to their objective quality; how much of it was due to their idiosyncratic fit; or how much of it, if any, was rooted in the ownership of the idea. To focus our test on the ownership part of the Not-Invented-Here bias, we needed to create a situation in which neither objective quality or idiosyncratic fit could be the driving force. (This, by the way, does not mean that the other two forces do not operate in the real world—of course they do. We only wanted to test whether the ownership of ideas is another force that can brin
g about overvaluations.)
To that end, we set up our next experiment. This time, we asked each of our participants to examine and evaluate six problems—the three that we used in the first experiment and three additional ones (see the list of the problems and the proposed solutions on the next page). But this time, instead of having some people take the role of creators and others the role of noncreators, we asked everyone to participate in both conditions (a within-participant design). Each participant evaluated three of the problems along with our proposed solutions, which put them in the role of noncreators. For the remaining three problems, we asked participants to come up with their own solutions and then evaluate them, which means that, for these three solutions, they were in the role of creators.*
Up to this point, the procedure sounds basically the same as the first experiment. The next difference was the important one for teasing apart the various possible explanations. We wanted the participants to come up with solutions on their own so that they felt ownership of them, but we also wanted them to come up with the exact same solutions that we came up with (so that better ideas or an idiosyncratic fit could not play a role). How could we achieve this feat?
Before I tell you what we did, take a look at the six problems and our proposed solutions on the next page. Remember that each participant saw only three of these problems with our proposed solutions and came up with solutions for the remaining three.
Problem 1: How can communities reduce the amount of water they use without imposing tough restrictions?
Proposed solution: Water lawns using recycled gray water recovered from household drains.
Problem 2: How can individuals help to promote our “gross national happiness”?
Proposed solution: Perform random acts of kindness on a regular basis.
Problem 3: What innovative change could be made to an alarm clock to make it more effective?
Proposed solution: If you hit snooze, your coworkers are notified via e-mail that you overslept.
Problem 4: How can social networking sites protect user privacy without restricting the flow of information?
Proposed solution: Use stringent default privacy settings, but allow users to relax them as necessary.
Problem 5: How can the public recover some of the money wasted on political campaigning?
Proposed solution: Challenge candidates to match their ad spending with charity contributions.
Problem 6: What’s one way to encourage Americans to save more for retirement?
Proposed solution: Just chat around the water cooler with colleagues about saving.
For each of the three problems for which participants had to come up with their own solutions, we gave them a list of fifty words and told them to use only these words to create their proposed solution. The trick was that each list was made of the words that made up our solution to that particular problem and several synonyms for each of these words. We hoped that this procedure would give the participants the feeling of ownership, while guaranteeing that their answers would be the same as ours.
For example, look at the list of possible words for answering the question “How can communities reduce the amount of water they use without imposing tough restrictions?” on the next page.
If you look closely at this list, you may notice another trick. We put the words that made up our proposed solution at the top of the list (water lawns using recycled gray water recovered from household drains), so participants saw those words first and consequently were more likely to come up with the same solution themselves.
We compared the value participants attributed to the three solutions we gave them with the three solutions that they “came up with.” Again, we found that participants appreciated their own solutions more. Even when we could not attribute the increase in perceived brilliance of the ideas to objective quality or to their idiosyncratic fit, the ownership component of the Not-Invented-Here bias was still going strong. At the end of the day, we concluded that once we feel that we have created something, we feel an increased sense of ownership—and we begin to overvalue the usefulness and the importance of “our” ideas.
Now, choosing a few words from a list of fifty to generate an idea is not very difficult, but it still takes some effort. We wondered if even less effort could make people think that an idea is theirs—the equivalent of Sandra Lee’s Semi-Homemade in the domain of ideas. What if we simply gave people our one-sentence solution but in a mixed word order? Would the simple act of reordering the words to form the solution be enough to make people think the idea was theirs and consequently overvalue it? For example, consider one of the problems we used:
Problem: How can communities reduce the amount of water they use without imposing tough restrictions?
Would New York Times readers be less impressed with the solution if it were written in a meaningful order and they were just asked to evaluate it? What would happen if we gave participants the same solution in a jumbled order and asked them to rearrange it into a grammatically correct sentence?
Here’s the solution written with words in a meaningful order:
Proposed solution: Water lawns using recycled gray water recovered from household drains.
And here’s the solution written with words out of order:
Words for the proposed solution: lawns drains using gray recycled recovered water household water from.
Did the jumbling make a difference? You bet! As it turned out, even reordering the words was sufficient for our participants to feel ownership and like the ideas better than the ones given to them.
Alas, we also discovered that Mark Twain was right.
A Negative Current
Now, you might ask, “Aren’t there areas—like scientific research—where the all-too-human preference for one’s own ideas takes a backseat? Where an idea is judged on its objective merits?”
As an academic, I wish I could tell you that the tendency to fall in love with our own ideas never happens in the clean, objective world of science. After all, we like to think that scientists care most about evidence and data and that they all work collectively, without pride or prejudice, toward a joint goal of advancing knowledge. This would be nice, but the reality is that science is carried out by human beings. As such, scientists are constrained by the same 20-watt-per-hour computing device (the brain) and the same biases (such as a preference for our own creations) as other mortals. In the scientific world, the Not-Invented-Here bias is fondly called the “toothbrush theory.” The idea is that everyone wants a toothbrush, everyone needs one, everyone has one, but no one wants to use anyone else’s.
“Wait,” you might argue. “It is very good for scientists to be overattached to their own theories. After all, this could motivate them to spend weeks and months in small laboratories and basements laboring over boring, tedious tasks.” Indeed, the Not-Invented-Here bias can create a higher level of commitment and cause people to follow through on ideas that are their own (or that they think are their own).
But as you’ve probably guessed, the Not-Invented-Here bias can also have a dark side. Consider a famous example of someone who fell too deeply in love with his own ideas and the cost associated with this fixation. In his book Blunder, Zachary Shore describes how Thomas Edison, the inventor of the lightbulb, fell hard for direct current (DC) electricity. A Serbian inventor named Nikola Tesla came to work for Edison and developed alternating current electricity (AC) under Edison’s supervision. Tesla argued that unlike direct current, alternating current could not only illuminate lightbulbs over greater distances, it could also power gigantic industrial machines using the same electrical grid. In short, Tesla claimed that the modern world required AC—and he was right. Only AC could provide the scale and scope needed for extensive use of electricity.
Edison, however, was so protective of his creation that he dismissed Tesla’s ideas as “splendid, but utterly impractical.”7 Edison could have had the patent for AC since Tesla had worked for him when he invented it, but
his love for DC was too strong.
Edison set out to discredit AC as dangerous, which indeed at the time it was. The worst that could happen to anyone who touched a live DC wire would be a powerful shock—jolting, but not lethal. Touching a live wire running AC, on the other hand, could kill instantly. The early AC systems of the late nineteenth century in New York City were made up of crisscrossed, overhanging, exposed wires. Repair workers had to cut through dead lines and reconnect faulty ones without adequate safeguards (which modern systems now have). Occasionally, people were electrocuted by alternating current.
One especially horrific case occurred on the afternoon of October 11, 1889. Above a crowded intersection in midtown Manhattan, a repairman named John Feeks was cutting through dead wires when he accidentally touched a live one. The shock was so intense that it cast him into a net of cables. The conjunction of charges ignited his body, sending streaks of blue light from his feet, mouth, and nose. Blood dripped down to the street below as onlookers gaped in horrified wonder. The case was precisely what Edison needed to bolster his charges about AC’s danger and thereby the superiority of his beloved DC.
As a competitive inventor, Edison was not about to let the future of direct current be dictated by chance, so he started a big public relations campaign against alternating current, attempting to generate public fear about the competing technology. He initially demonstrated the dangers of AC by directing his technicians to electrocute stray cats and dogs, and used this to show the potential risks of alternating current. As his next step, he secretly funded the development of an electric chair based on alternating current for the purposes of capital punishment. The first person ever to be executed in the electric chair, William Kemmler, was slowly cooked alive. Not Edison’s finest moment, to be sure, but it was a very effective and rather frightening demonstration of the dangers of alternating current. But despite all of Edison’s attempts to foil it, alternating current eventually prevailed.