Innovators Need To Both Explore And Exploit

The innovation literature often gets bogged down in the so-called dual operating system that so troubles organizations as they attempt to exploit existing knowledge while also striking new territory.  On one hand, you have more iterative improvements of what is already known to the organization, whereas on the other you are exploring the unknown.

New research from Harvard Business School highlights that this exploration and exploitation divide may not be quite as clear cut as we previously thought.  Indeed, it argues that breakthrough innovations can often be a result of both exploration and exploitation approaches.

“For instance, one of the most important inventions of the 20th century, the transistor, was created at the Bell Labs through a process that exhibited the classic hallmarks of both exploration (e.g. working on the frontiers of known science, experimentation with unprecedented design concepts, unpredictable paths forward, and failure) and exploitation (cumulative learning, incremental refinement and improvement of materials’ purity, and iterative experimentation),” the researchers explain.

A dual approach

The researchers test this hypothesis by analyzing patent data from over 2,500 organizations between 1975 and 2005. In total, this covers over 1.3 million patents. Through their analysis, they aim to understand the “distance” in theoretical knowledge between any given patent and the patent portfolio for the firm in the prior year. The closer the proximity the more it would suggest the firm is exploiting rather than exploring.

The data reveals that firms often begin with a period of exploration, with this then followed by a period of exploitation.

“Specifically, when firms innovate by first venturing into a distant area of the theoretical knowledge space and then focus their activities on that area and iterate within it, they are more likely to produce a breakthrough, than when following a different search process,” the researchers explain.

Breakthrough innovations

What’s more, the data suggest that the shape of the search trajectory used by the firm and its intensity have a big impact on the likelihood that the invention will be a breakthrough one.

As such, it appears that the best innovators begin their journey by exploring unfamiliar territory, with this territory then becoming a focal point for subsequent exploitation, such that the previous unfamiliar territory becomes familiar. It’s ultimately the exploitation of this now-familiar body of knowledge that leads to breakthrough innovations.

Of course, achieving this dual approach is far from easy as exploration and exploitation require very different skills, structures, processes, and cultures. The researchers contest the notion that the best method is to place exploration and exploitation in distinct units.

“To the extent breakthroughs require both, then understanding how these seemingly contradictory capabilities can be integrated looks to be an organizational challenge well worth understanding in future research,” the researchers conclude.

Facebooktwitterredditpinterestlinkedinmail