DE&I Metrics: Are they useful to achieving DE&I Goals?

Given the recent uptick in companies disclosing DEI metrics, how effective is this disclosure to actually moving the needle? How should the disclosure of DE&I metrics fit into overall DE&I strategy?

The push for companies to publicly disclose diversity metrics for accountability and potential incorporation into shareholder evalutation is not new, despite the renewed push for transparency following the murder of George Flloyd and many other unarmed African Americans in 2020 which resulted in a summer of protest and push for racial justice. Beginning in 2014 large tech companies like Apple, Facebook, Google and Microsoft all started publishing annual reports; however, for DE&I proponents, there has been far too little progress in the disclosed metrics.

In the recent article How to Best Use Data to Meet Your DE&I Goals, Siri Chilazi and Iris Bohnet argue that disclosing DE&I metrics can only bring about changes in behavior if accompanied with the following four actions:

  1. “Present diversity data in a way that is simple, salient, and comparable”

Their first argument is that DE&I metrics have to be easy to understand, easy to follow the connection to results, and easy to compare with other companies. There are three important points in this argument: 1. If metrics are not easy to understand and digest, then they have no chance of changing behaviors. 2. Metrics must be salient in order to be actionable. 3. If metrics are not easily comparable with other companies’ metrics, then users will be forced to look at the metrics in a vacuum, rendering them relatively useless for decisions. 


  1. “Leverage diversity data to empower the right people to act.” 

Their second argument hinges on the idea that within corporations, DE&I metrics are frequently tracked by HR, D&I and People Operations. Most front-line workers have no visibility into the metrics, and senior leaders only see them during scheduled reports or dashboards, in more advanced organizations.


  1. “Set diversity goals to create accountability and increase follow-through.”

The third argument discusses a typical DE&I understanding, which is what is not measured within an organization does not count, but goes further to say that disclosing tangible goals results in accountability and also pushes the organization to make cultural changes to reach those stated goals. 


  1. “Leverage diversity data to shift social norms around DE&I.” 

Their final argument dispels an idea held within the tech world that disclosure reinforces the idea of exclusion by saying that DE&I metrics have shifted much more quickly after public disclosure began. 



These four arguments are really about how to tailor the data to be actionable by the following four stakeholders: Shareholders / Public, General Employee Population, Employees Responsible for Execution, and Senior Leadership. 

While the argument that actions must follow data aligns with the typical business strategy flow, when it comes to DE&I metrics, companies would benefit by starting with the ideal state of DE&I and their culture and then creating an action plan regardless of prioritizing those actions that will shift metrics. 

This is not an argument to stop using DE&I metrics, and it is not an argument to say there is no power in publicly discussing DE&I metrics and goals. It is an argument that not all of the actions to create a more equitable culture can be tangibly measured, and if companies are so focused on producing results in the form of metrics and justifying DE&I program spend with these metrics, then actions that could truly produce a safer, more just workplace for minorities could be deprioritized. 

A very tangible example is around mentorship, learning and development and ERG efforts. These programs alone might not have a direct impact on DE&I metrics, and thus could be deprioritized or seen as expendable, but could be the most important factors in the success of minority groups within a workplace. 

Overall, the article by Chilazi and Bohnet clearly articulates how to ensure that DE&I metrics are salient and actionable for a variety of stakeholders, but needs to be put into overall DE&I strategy setting context. 


Making Sense of the Virtual Environment: The Power of People Analytics for Reflecting on the Pandemic


Ford: Brick & Mortar Icon Transforms Workforce Home Office Capabilities

Student comments on DE&I Metrics: Are they useful to achieving DE&I Goals?

  1. Hi Christina,

    Thank you for bringing up the very important topic of DE&I and the challenges around metrics. I absolutely agree that there should be an agreed business standard that allows all companies to share and compare their results, and by making the information visible, to allow for more company-wide initiatives to promote diversity.

    As a proponent of women / Latina representation in corporate America, I’m always faced with the question of what should be our markers in the DE&I work. Is having the population of employees perceptually representative to the country’s population the goal? This would mean that based on the 2010 census (while we wait for the 2020 census report) and despite being quite far from this goal, Hispanic women should constitute 8-9% of the employees and that should be enough. Or for African American women, 6-7%, and so forth. What happens once we reach that figure? And what considerations need to be made for different industries or disciplines, i.e. engineering vs. HR, project management vs. marketing, each of which tend to have greater male vs. female representation (depending on the industry as well). Either way, having access to these metrics can help guide education and employment initiatives to push forward a fairer playing field.

    I’m a big fan of Iris and her behavioral science insights, and hope more companies take these recommendations to heart.

  2. Interesting insight on this growing aspect of DE&I efforts, Christina. I would push the envelope one step further to consider how easily gamed the numbers can be. For example, my old firm was forced to disclose minority representation. Instead of being transparent about the terribly awful numbers, they decided to group together all possible minority groups to compare against the white majority. The numbers were skewed due to the multiple office locations in Asia that counted nearly all employees as minority (although they were majority members in their countries). They hid the near zero numbers of African American and Hispanic employees in the U.S. offices.

    My point is that we can create a false sense of progress with numbers and I agree that there are measures that can go unaccounted for in the search for quantification.

Leave a comment