This article is about how investing in R&D is fundamental. A research shows that generalists are better to identify valuable opportunities.

Article by Frank Nagle and Florenta Teodoridis for Harvard Business Review | 28 September 2020

Innovation is an increasingly important source of competitive advantage for companies: Over the last 40 years, organizations have increased their expenditures on R&D as a percentage of all expenses by 800%. However, increased R&D budgets alone do not always lead to real innovation. The types of researchers a company hires plays an important role in whether it will succeed in leaping ahead of competition. But what types of researchers are most valuable?

Conventional wisdom recommends hiring researchers with deep expertise in a single, desirable field, such as artificial intelligence, neuroscience, aerospace. However, our recent research shows that this alone is not sufficient. Hiring more researchers with diverse areas of expertise — “jacks of all trades” — is increasingly necessary for companies to successfully compete.

These researchers are often viewed as having lower abilities and skills than their more focused counterparts, in keeping with the adage “a jack of all trades is a master of none.” After all, they lack specialists’ depth of knowledge. Yet we found these researchers’ diversity of knowledge and interests serves an important function in innovation, and is not a sign of low ability or skill.

The Tradeoffs of Specialists vs. Generalists

At its core, innovation is the result of creatively combining various kinds of knowledge. However, the scope of such combinations can vary greatly: Advantages can be won from exploiting existing competencies for incremental gain, but also from dynamic combinations that generate radical change. 

Research has shown that it is often this activity of broader exploration that leads to the greatest positive impact for companies. However, it is unrealistic to expect most organizations to maintain a broad set of employees across many domains of knowledge in hopes of identifying powerful combinations. What is the alternative?

Many studies document advantages and disadvantages of hiring specialized or diversified researchers, and the trade-off between the depth and breadth of knowledge in particular. Specialized researchers have the advantage of deep knowledge in their area of expertise, but narrow breadth across other domains. Conversely, diversified researchers have the advantage of breadth, but their knowledge in each domain is more shallow.

Our research shows that individuals with diverse research portfolios — generalists — are more likely to identify valuable opportunities from new fields. While depth of knowledge is needed to successfully execute on an idea, we found that generalists were better at making connections across specialties and discovering promising combinations. Absent the discovery step, the ability to execute is moot.

Why Generalists Excel

To get as close to the gold standard of a randomized, controlled research experiment, we exploited a quasi-natural experimental setting: the unexpected hacking of the Microsoft Kinect. Microsoft Kinect, an add-on for the Xbox 360 video game system, was launched in November 2010 in the video game market.

Technology enthusiasts and scholars hacked Kinect in a coordinated effort to repurpose it as a motion-sensing research technology in fields ranging from artificial intelligence, robotics, and virtual reality to paleontology, education, health care, music, cinematography, market research, and advertising. It was a classic example of the widely accepted view that technology facilitates access to new knowledge. We used this unexpected event to estimate the propensity of diversified and specialized researchers to successfully engage with Kinect in research.

As our dataset, we used a sample of over 180,000 researchers in the engineering domain with varying degrees of research diversification, as evidenced by their research publication histories. Our work showed that generalists (the top 25 percent most diversified researchers) were 3.1 times more likely to use Kinect in their research within the first four years compared to specialists (the bottom 25 percent of diversification) of similar ability.

Further, and perhaps more importantly, our study also shows that the research generalists produced was 3.8 times more likely to be highly cited, indicating their work was more impactful. How were they so successful? In a related study, one of us (Florenta) found that generalists achieved success by engaging in collaboration with larger and more diverse teams of specialists. This gave them access to the deep knowledge that they lacked, while leveraging their ability to combine information from across disciplines and lead to impactful innovations.

Building Innovative Teams

Our research offers two important insights for managers attempting to build highly innovative R&D teams. First, successful exploratory innovation is increasingly a team effort that requires both diversified researchers who specialize in scanning available and new knowledge to identify combinations with a high probability of generating impactful innovations, and domain specialists who can exploit the identified opportunities for exploratory innovation.

This balancing act requires that organizations consider the strategic mix between generalists and specialists when hiring. To achieve the balance, organizations need to devise approaches to identify not only high-ability specialists, but also high-ability generalists. Therefore, in the short-run, organizations can evaluate researchers’ innovation portfolios with respect to both impact and spread across knowledge domains. High-ability generalists can also be identified by their wide network of high-ability specialist collaborators.

Second, managers need to navigate the process of hiring generalists with care, as existing labor market signals and procedures favor identifying specialists and penalizing generalists. Rewards, such as prizes, grants and promotions, are typically awarded to successful specialized researchers. The same set of incentives weed out generalists, regardless of whether they’re of high or low ability. Given that the current incentives are specialist-biased, individuals who would make great generalists are instead encouraged to specialize, leading to an undersupply of generalists. Thus, in the long-run, organizations should consider incentivizing individuals to become generalists. In our research, we find that employer incentives are frequently more important to a researcher becoming diversified than the researcher’s personal preferences.

Jack of All Trades and Master of Knowledge

While combining knowledge from different domains has long been recognized as a cornerstone of competitive advantage, many companies continue to only hire specialists in an attempt to achieve this goal. Our research shows that this strategy is costly and outdated for three reasons.

First, it is not typically feasible to hire specialists with expertise in a wide set of domains, and so companies end up selecting the most relevant domains to keep costs manageable. Second, as knowledge continues to be discovered, the body of knowledge available continues to increase. Along with it, the set of domains one can specialize in expands, as does an organization’s need to frequently update their roster of specialists.

Finally, the updates would entail an increase in the number of specialists necessary to continue to cover the same fraction of the entire body of knowledge available. These strategies are unsustainable.

In the end, not all companies need to conduct exploration for radical innovations to obtain competitive advantage. However, those who do should give serious consideration to increasing their hiring of generalist researchers. Contrary to the adage describing generalists as a Jack of All Trades and Master of None, our research shows that, in innovation, these individuals are instead a Jack of All Trades and Master of Knowledge.