Download the report

Search 2.0


A definitive guide to more inclusive outcomes

Search 2.0


A definitive guide to more inclusive outcomes

Search 2.0


A definitive guide to more inclusive outcomes


Protecting the brain’s processing power has been important to our evolutionary history. The brain relies on “shortcuts” (intuitions, gut feelings, common sense, muscle memory, instincts, etc.) to conserve processing power. This happens through instantaneously linking situations to patterns from memory and applying stored solutions to upcoming decisions. People can misunderstand the nature of intuition and portray it as irrational. It isn’t. It is a rational but unconscious process of making decisions that suffices for most daily decisions. Only a small minority of decisions deploy the conscious processing power of the brain.

Both our intuitions and analyses are more often than not correct and are essential to our survival and thriving as a species. However, both are also prone to errors. The study of these errors is the field of cognitive biases.

There are close to 200 known cognitive biases. The most common and relevant biases that we have observed in senior talent decision settings are listed below alphabetically. You may want to add to or shorten the list based on your own experiences. They don’t all apply to every situation, and in some cases, they may even counterbalance one another. These are effectively the higher-order processing flaws that could impact a range of underlying decisions that can negatively affect DEI outcomes. 

Ambiguity bias

Authority bias

Anchoring bias

Affinity bias

Attractiveness bias

Attribution bias

Confirmation bias

Conformity bias

Effort justification bias

Egocentric bias

Halo bias

Horn bias

Objectivity bias

Present bias

Status-quo bias

Zero-risk bias

The tendency to avoid options for which the probability of a favorable outcome is unknown (e.g., viewing diverse candidates as higher risk).

The tendency to attribute greater accuracy to the opinion of an authority figure unrelated to its content (e.g., the boss/expert knows best).

The tendency to disproportionately rely on one piece or a narrow set of information, usually received first or early, to guide decisions (e.g., receiving strong positive or negative commentary on someone).

The tendency to be naturally drawn toward individuals with similar characteristics, backgrounds and interests (e.g., alumni of the same university).

The tendency to associate more favorable characteristics with those who may look or dress in conventionally desirable ways.

The tendency to overemphasize personal factors and underestimate situational factors when evaluating an individual (e.g., ignoring macro situations and benchmarks while evaluating performance, positively or negatively).

The tendency to selectively search for or interpret information in a way that confirms one’s preconceptions (e.g., in final referencing for candidates).

The tendency to adapt our own opinions to fit with those of a group.

The tendency to attribute greater value to someone’s success if you have had a role in supporting them (e.g., when favoring internal candidates or those you have mentored or worked with in the past).

The tendency to place a higher emphasis on and belief in the accuracy of one’s own perspectives than those of others.

The tendency for positive aspects in a profile to spill over and positively influence the view of other unrelated aspects (e.g., “They went to Harvard, they must be good.”).

The tendency for negative aspects in a profile to spill over and negatively influence the view on other unrelated aspects (e.g., not hiring someone who worked for a company that had bad press).

The tendency to believe that one is more objective and unbiased than others.

The tendency to favor lower immediate payoffs relative to greater later payoffs (e.g., “We cannot wait for a better candidate; we need to solve this now.”).

The tendency to favor the default or incumbent situation in comparison to making a change (e.g., “They are not performing; let’s give it six more months.”).

The tendency to prefer reducing a small risk to zero over a greater reduction in a larger risk (e.g., refusing to accept any compromise in prior experience at the expense of the much greater overall benefits another candidate could bring).

Pathway to reducing Biases

About the Author

Your Egon Zehnder Team

Special thanks to colleagues across the Firm who contributed their knowledge and time to this project:

Michael Ensser, Edilson Camara, Helen Crowley, Cagla Bekbolet, Abed Saleh, Fiona McGauchie, Yasushi Maruyama, Chie Iida, Neil Waters, Yan Geng, Namrita Jhangiani, Shilpa Rangaswamy, Lena Kilee, So-Ang Park, Kine Seck Mercier, Sandra Garcia, Ingrid van den Maegdenberg, Marike Kuin, Pam Warren, Dede Orraca-Cecil, Cynthia Soledad, Angela Pegas, Fabio Nunes, Gizem Weggemans, Mark Longworth, Charlotte Wright​ , Obinna Onyeagoro, Paul Havranek, Christian Schmidt, Loula Lefkaritis, Alessandra Tosi, Carol SingletonSlade, Claire Thomas, Anthony Cavanough, Ashley Summerfield, and Andrew Roscoe.

Global Diversity Specialist:​ Katrin Sier
Research:​ Ryan Hoffmann, Raminder Kaur, Kathrin Heinitz, Nathia Pratista
Editorial Team:​ Cheryl Martel, Luisa Zottis
Design Team: Richard Khuptong, Markus Schuler, Kamaljit Marwaha, Vijayakumar Shanmugamani, Dapinder Pal Singh Bahl
Digital Team:​ Amadeu Porto, Becky Neems, Joanna Scheffel, Aditya Gupta, Arnab Kar
Executive Assistant:​ Hannah Hughes