Rogers' Diffusion of Innovations

Everett M. Rogers defined diffusion as the process by which an innovation is communicated through certain channels over time among the members of a social system. His definition names four elements: the innovation, the communication channels, time and the social system.

Rogers built the theory by synthesizing a large body of studies, starting with rural sociology research on how farmers adopted hybrid seed corn in the 1940s. The first edition of Diffusion of Innovations appeared in 1962; the fifth edition, published in 2003, is the one most often cited.

When cumulative adoption is plotted over time, it typically follows an S-curve: slow at first, then rapid, then leveling off.

Rogers' diffusion of innovations therefore treats adoption as a social process, not only an individual decision.

Rogers divided adopters by when they adopt, measured in standard deviations from the average adoption time. On a normal curve this gives approximate shares:

The shares are ideal types derived from the curve, not counts from any one study.

Rogers' adopter categories therefore describe relative timing, not fixed personality types.

Innovators try new ideas first, but peers often see them as unusual, so their example carries limited weight. Early adopters are different: they are embedded in the local system and respected, and others watch them before deciding. Rogers found that this group has the highest degree of opinion leadership.

An opinion leader is someone whose views informally shape how others think and act. Opinion leadership comes from standing and connection, not job title.

Most people do not adopt because of formal communication alone. They adopt after seeing someone like them use the innovation and benefit from it.

Rogers' diffusion of innovations therefore puts early adopters at the center of any plan to spread a change.

Rogers identified five perceived attributes that affect how fast an innovation is adopted: relative advantage, compatibility with existing values and practices, complexity, trialability and observability.

Larchfield Logistics is introducing handheld scanning in its warehouses. The change team:

1. identifies respected shift leads in each warehouse, not only the volunteers who like new devices;
2. runs a short trial with them, so others can watch (trialability, observability);
3. collects the time each shift lead saves on stock checks and shares it in team briefings (relative advantage);
4. keeps screen labels close to the existing paper forms (compatibility, lower complexity).

Rogers' diffusion of innovations therefore gives a change manager both a sequence of whom to engage and a checklist of what slows adoption.

Diffusion of innovations rests on hundreds of studies across fields, which gives it a broader evidence base than most change models. Its limits are known, and Rogers discussed several himself.

• Pro-innovation bias. The theory tends to assume that an innovation should be adopted, so non-adoption looks like a failing rather than a sound judgment.
• Voluntary adoption. Many workplace changes are authority decisions. People must adopt, so timing reflects the rollout more than personal choice.
• Labels. Calling people "laggards" can stigmatize groups whose caution is reasonable.
• Context. A systematic review of innovation in health services by Trisha Greenhalgh and colleagues found that adoption in organizations depends heavily on context, readiness and implementation, not only on adopter traits.

Rogers' diffusion of innovations is therefore a useful map of how adoption spreads, but not a prediction for any single change.