Countless psychology studies have proven that we tend to like people who look and/or think like us. Peter Salovey, Dean of Yale College and a Professor of Psychology, taught my “Introduction to Psychology” class in college. On the first day of class, he asked, “Do you want to know whom you are going to marry?” The 18-year-olds were excited to learn the answer. Professor Salovey says, “Look at the person next to you.”
Many students are disappointed at the type (and gender) of the person they are sitting next to. However, Professor Salovey’s point is still valid. He is teaching that we tend to place ourselves next to and in the same environment as those who resemble ourselves, i.e., selection bias . We are thus more likely to befriend those people, we are more likely to develop strong ties to them, and even eventually to marry one of those people. More generally, people are disproportionately likely to develop ties that are homogeneous by values (both conservative), interests (both golfers), affiliation (members of the same Rotary club), and participation (take a hip-hop dance class together).
However, our tendency towards the familiar can jeopardize the process of team building in a company. Often, when startup founders first assemble teams, they hire familiar people – either people they know, or people like them. In a rigorous study, Howard Aldrich, Martin Ruef, and Nancy Carter found that entrepreneurs pick founding team members based on trust and people who are similar to them, not based on identifying people with the most-needed skills (see the 2003 paper).
The problem is that you have chosen the people you know for many reasons: because you were randomly assigned roommates and ‘get along’, you like the same football team, or they married your sister. Those are good reasons to be friends, but poor reasons to hire someone. When you want to hire someone, you should strictly focus on hiring the best possible person for the open job. It is statistically highly unlikely that you already know the best qualified person for a given job at your company. So you need to go out and systematically recruit that person.
In addition, hiring like people jeopardizes one of the most basic tenets of a successful, highly adaptable team: diversity. This is not just politically correct cant, but backed up by the academic research on this topic. Diversity matters along all axes: the conventional ones of race & gender & religion, but also the less conventional ones of socio-economic status, functional background, marital status, etc.
Most startup companies address a wide market and thus cater to a diverse audience, and these types of companies benefit from a team that brings myriad experiences and viewpoints to the table. Some companies, on the other hand, address very specific types of audiences: the key, here, is to remember that an understanding of the core customer is just as important as product development / technical expertise as an asset to the company.
This Inc. article discusses our tendency to hire people who look and think like we do. This is dangerous, the author argues, to the development of an adaptable, thoughtful culture. She writes, “The company in which there is no conflict is the one where there’s no debate and precious little thinking. The reason you need people not like you is because they will spark argument and dissent. Because they don’t share your worldview, their ability to spot risks or greater opportunities is immensely valuable. You need to find these people, cherish them—and not turn them into clones.”
A lot of startups are looking for people who ‘fit into the culture’. A healthy culture can tolerate some diversity of background. Core to a successful culture is a culture of high performance, and the highest performing people may well come from a different background than you do.
(Thanks to Joan Xie for help researching this.)
 Matthijs Kalmijn, “INTERMARRIAGE AND HOMOGAMY: Causes, Patterns, Trends,” Annual Review of Sociology 24 (Aug 1998), 395-421.
 Peter V. Marsden, “Core Discussion Networks of Americans,” American Sociological Review 52 (1987): 122-131.