Not-Invented-Here Syndrome: Why Outside Ideas Get a Harder Hearing

Groups judge ideas partly by where they come from. An outside proposal arrives without the shared history that makes an insider's proposal legible: the evaluator does not know what was tried or what the author assumed. Adopting it also costs something that adopting your own does not. It says the group did not solve the problem itself, and where teams compete for budget, headcount or standing, that admission has a price.

Tanya Menon and colleagues showed in 2006 that the price depends on who the outsider is. Managers in their studies devalued knowledge from an internal rival, another unit of the same company, and were more willing to take the same knowledge from an external rival, a competing firm. Learning from a colleague down the hall threatens status in a way that learning from a competitor does not. Earlier, Menon and Pfeffer had found in 2003 that outsiders' knowledge is sometimes preferred to insiders', because an outsider's ideas are scarce and the insider's are easily scrutinized. The bias is not a blanket suspicion of everything foreign; it is sharpest for ideas from close rivals and for ideas whose adoption would cost the group face.

Katz and Allen's account of the syndrome was about isolation rather than rivalry. Long-tenured project groups in their study talked less with colleagues outside the group, and with the outside world in general, than newer groups did, and their managers rated their performance lower. The authors' interpretation was that a stable group comes to believe it holds a monopoly of knowledge in its field and stops importing ideas, to its own detriment. Antons and Piller, reviewing the literature in 2015, defined the syndrome as a negative attitude toward external knowledge, held by the group, that acts as a decision bias.

The effect is easy to overstate. The Katz and Allen data are a correlation between tenure and rated performance in one laboratory, and Takahashi and Inamizu argued in 2012 that the famous curve comes from heavily smoothing a scatter plot with no clear trend, and that what Katz and Allen called the NIH syndrome was a tenure effect, not a measured rejection of outside ideas. The attitude itself has been demonstrated in controlled studies such as Menon's; its size in any particular organization is an empirical question, not a given.

Ralph Katz and Thomas Allen published "Investigating the Not Invented Here (NIH) syndrome: A look at the performance, tenure, and communication patterns of 50 R&D project groups" in R&D Management in 1982. The setting was the research and development facility of one large American corporation, where all 345 technical professionals were assigned to one of 61 project groups; complete data were obtained for 50 of the groups. The authors defined the syndrome as the tendency of a project group of stable composition to believe it possesses a monopoly of knowledge in its field, and so to reject new ideas from outsiders to the likely detriment of its performance.

Group tenure was the average time a project's members had worked together. Communication was measured by asking every professional, on one randomly chosen day each week for fifteen weeks, whom they had talked to about technical matters. Performance was rated by the laboratory's seven department managers and two research directors on a seven-point scale, with each project scored by about five raters and the scores averaged.

Plotting rated performance against average tenure, the authors reported that performance rose over the first year and a half, held steady for a time, and had declined noticeably by five years. The longest-tenured groups also talked least, among themselves and with the critical external sources of information in their field. Katz and Allen read the decline as the groups cutting themselves off, and suggested rotating members and giving groups new problems.

The paper shows an association between group tenure, communication and rated performance in one laboratory; it does not present any group rejecting a specific outside idea, and the performance measure is a manager's rating. The experimental evidence that people devalue knowledge by its origin came later, in studies such as Menon, Thompson and Choi's in 2006, where the content was held constant and only the stated source changed, which is the design the demo on this page borrows.

The rows below are illustrations of where an idea's origin colors its reception, not findings from the studies cited on this page.

Because the bias rides on origin, most of the counters work by separating the judgment of content from the knowledge of where the content came from, or by changing what adoption costs.

• Read first, attribute later. Circulate proposals for a first reading with the author removed, and record the first judgment before the name is revealed.
• State the evaluation criteria before the proposal arrives. A plan judged against written criteria is harder to pick apart on whatever grounds happen to be available.
• Make adoption cheap in status. Credit the team that adopts an outside idea as much as the team that produced it, and say so in advance; Menon's studies suggest the threat to standing is a large part of the resistance.
• Put a price on building your own. A written estimate of the replacement makes the comparison fair.
• Keep the group's edges open. Katz and Allen recommended rotating members and importing new problems for long-tenured teams; a group that talks to nobody outside has no way to notice what it is missing.

Several biases concern where an idea comes from, and they overlap at the edges. The not-invented-here syndrome is specifically the group-level preference for its own output over outside output.

• It is not reactive devaluation, in which a proposal is rated worse because an adversary in a negotiation made it. That effect concerns a counterpart's offer and the suspicion of concession; the syndrome concerns any outside origin, rivalry or not, including a tool nobody is bargaining over.
• It is not in-group favoritism, the general tendency to favor members of one's own group. The syndrome is the special case about knowledge and ideas, and it has its own extra ingredient: the status cost of admitting the group did not think of something.
• It is not the IKEA effect, the overvaluation of things you built yourself. That effect is about effort and ownership of a finished object; the syndrome operates before anything is built, at the point of deciding whose idea to use.
• It is not the Semmelweis reflex, the rejection of new evidence because it contradicts an established view. The syndrome rejects ideas for coming from outside, even when they contradict nothing.