Pro-Innovation Bias: Why New Things Are Assumed to Belong to Everyone

The bias begins with a selection. Diffusion research grew up studying innovations that had succeeded: hybrid seed corn in Iowa, a new antibiotic among physicians, practices promoted by agricultural extension agencies. Rogers pointed out that such studies are usually sponsored by the organization promoting the innovation, are conducted after the innovation has spread, and sample the people who adopted it. An innovation that was rejected, abandoned after a trial or quietly altered by its users leaves fewer traces and attracts no sponsor, so it is rarely studied.

From that sample a set of assumptions followed, and Rogers listed them: that the innovation should be adopted by all members of the social system, that it should be adopted faster, and that it should be adopted as designed, without re-invention or rejection. He noted that the bias was seldom stated, only implied, which made it harder to see. Its consequences were specific. Researchers neglected ignorance of innovations, underemphasized rejection and discontinuance, overlooked re-invention, and did not study programs designed to prevent the spread of harmful innovations.

The bias has a companion. When adoption falls short, a researcher who has assumed the innovation is good has one place left to look for the cause: the people who did not adopt. Rogers called this the individual-blame bias, the tendency to hold the individual responsible for a problem rather than the system, and traced it to the same source. He did not argue that individual variables are never the right ones, only that they are rarely the whole explanation.

Outside research, the same assumptions appear whenever a launch is covered as if adoption were a matter of time. The inventor's claim that a product will replace the car and the investor's forecast of unprecedented growth each take the innovation's merit as given and treat the public's job as catching up. Some launches do reach nearly everyone, which is why the demo includes one. The bias is not in expecting success but in treating it as the only outcome worth planning for, and in reading every non-adopter as a laggard.

Pro-innovation bias is not an experimental finding. It is a criticism Everett Rogers made of his own field in Diffusion of Innovations, a book first published in 1962 and revised through a fifth edition in 2003, and in a 1976 article in the Journal of Consumer Research on what diffusion research had contributed to the understanding of new product adoption and where its background had limited it. Rogers traced recognition of the bias to his 1971 book with Floyd Shoemaker, and wrote in later editions that very little had been done to remedy it since.

His definition is the one quoted on this page: the implication of most diffusion research that an innovation should be diffused and adopted by all members of a social system, that it should be diffused more rapidly, and that it should be neither re-invented nor rejected. He gave reasons for it. Historically, the innovation that founded the field, hybrid corn in the 1943 study by Ryan and Gross, really was profitable for the Iowa farmers who adopted it, so the first great diffusion study had little reason to ask whether adoption was wise. Institutionally, diffusion research was funded by the agencies that promoted innovations, who had no interest in studying rejection. Methodologically, successful innovations are easy to study after the fact, because their adopters can be found; failed ones are not.

Rogers also set out remedies. Researchers could study an innovation while it was still spreading rather than only after success, could study innovations that were rejected, discontinued or re-invented, could investigate how the decision to diffuse an innovation was made in the first place, and could ask about adopters' motivations instead of assuming an economic one. And they could keep an open mind toward system-level explanations of non-adoption, in place of the individual-blame habit he described in the same chapter. Later editions of the book used cases such as the Dvorak keyboard, a design that did not spread despite its claimed advantages, and the failure of a water-boiling campaign in a Peruvian village, to show that merit alone does not diffuse an innovation.

Because the concept comes from a critique rather than an experiment, this page has no participant count or effect size to report, and the demo tests your forecasts against the public record of three launches. The three cases illustrate the assumption Rogers described; they are not drawn from his book.

The rows below are places where the assumption that a new thing belongs to everyone does its work. They are illustrations, not findings from the sources cited on this page.

Rogers's remedies were written for researchers, and they translate directly for anyone judging or promoting something new.

• Study the non-adopters as carefully as the adopters, and start from the possibility that the innovation does not fit them rather than that they have failed to understand it.
• Count rejections and discontinuances. An adoption rate without the number who tried the thing and stopped is half a picture.
• Look for re-invention. The people who changed the innovation to make it work are telling you what was wrong with it as shipped.
• Ask for the base rate. Of launches with this pitch and this backing, how many reached most people? The record for new products is mostly niches and failures, with a few exceptions that get all the coverage.
• Watch the direction of blame. When a shortfall in adoption is explained entirely by the character of the people who did not adopt, the individual-blame bias is at work, and the system and the innovation have been excused without a hearing.

Pro-innovation bias has close relatives that pull the same way and others that pull the opposite way.

• It is not appeal to novelty, the judgment that a newer thing is better because it is newer. That is an individual's error about quality; pro-innovation bias is an assumption about spread, that a good innovation will and should reach everyone, built into how institutions study and promote it. A person can judge an innovation's quality correctly and still assume everyone should have it.
• It is not the bandwagon effect, adopting something because others have. Bandwagon describes adopters following adopters; pro-innovation bias describes the observer who assumes the bandwagon is the whole story and the people not on it are the problem.
• It is not the opposite of status-quo bias or not-invented-here syndrome, the preferences for what one already has. Those are errors of the non-adopter; Rogers's point was that the observer's error is to see only those, and never the innovation's own misfit.