Download This Report[/button]

Introduction

At this point, the business case for diversity, equity, inclusion, and belonging (DEIB) is clear. Our own research (see Figure 1) shows the relationship between having a strong DEIB culture, and critical individual and performance outcomes.¹

Yet, for years, the representation of diverse populations in organizations improved almost imperceptibly.

Then we had a global pandemic and the rise of a social justice movement, sparked by the murder of George Floyd. Along with that came the heightened awareness that the pandemic was impacting diverse populations much more—particularly for women and people of color who were dropping out of the workforce at higher rates than other populations. As a result of this confluence of events, organizations began making big promises on DEIB in the summer of 2020.

When this happened, one of our first questions was how organizations would show that they’d made good—or at least made progress—on those commitments. While DEIB metrics measurements designed to understand DEIB—are the obvious answer, how to select, collect, use, and maintain those metrics is not so clear.

Thus, this research initiative on DEIB metrics and analytics was born. The first article in this series, “DEIB Analytics: A Guide to Why & How to Get Started,” provides leaders with a plan on how to begin using DEIB metrics and analytics. We’ve shared an 8-step guide with details on the actions and considerations that organizations need to take to effectively implement DEIB metrics.

This article: An essential guide to DEIB metrics

This report focuses more narrowly on the appropriate metrics and analytics for DEIB. We aim to provide DEIB leaders, people analytics practitioners, HR business partners, workforce planning and talent management leaders with:

1. A foundational understanding of the different metrics that can be used to measure and track their DEIB performance
2. Insights on how those different metrics might vary, depending on their org’s sophistication with DEIB and analytics

This article is based on a wide range of information, including our research on:

• People analytics technology²
• DEIB analytics³
• DEIB strategies⁴
• DEIB technology⁵
• A literature review of DEIB and analytics⁶
• Interviews with ~20 people analytics and DEIB practitioners

Our research focuses specifically on the people within an organization’s existing workforce. We know a number of other DEIB metrics exist that orgs should also consider, such as those which apply to their supply chain, community efforts, ESG (environmental, social, and governmental) requirements, etc. While critical, those areas are outside the scope of this report.

We would also like to mention that this report is the first of its kind, in that it attempts to provide a holistic look at all talent-related DEIB metrics. Any first try will miss some critical elements and we acknowledge this report may be incomplete. We invite you to share any suggestions, feedback, or additions you think appropriate by emailing us at [email protected].

The DEIB space is evolving quickly, and we will only make progress by putting out our best ideas and amending them quickly as new information becomes available. Thank you for being part of that process and pushing forward toward greater opportunities for all.

Defining DEIB

Let’s start our essential guide by defining our terms (see Figure 2).

Why are DEIB metrics & analytics important?

Some of the common reasons why leaders start to focus on DEIB metrics and analytics include:

• Creating a clear business case for DEIB
• Measuring the return on investment (ROI) of DEIB expenditures
• Tracking the impact of critical DEIB initiatives

In addition to these, a few more reasons why orgs should use DEIB metrics and analytics include:

• Busting myths or addressing anecdotes that may or may not be true
• Checking assumptions about DEIB
• Meeting consumer, investor, and employee expectations when it comes to progress on DEIB

While these are all good reasons to use DEIB data, one of the most compelling motivations for why DEIB is critical was articulated by one of our interviewees:

“Companies have been setting diversity goals for decades but have struggled with “goal-getting”—meaning the clear accomplishment of those goals—because of a lack of feedback and data to help them get after those goals every day. Without any feedback on progress, companies lose sight of the goals.”

—Phil Willburn, Head of People Analytics & Insights, Workday⁷

Why do orgs find DEIB data difficult to use?

Many leaders struggle to use DEIB data for reasons such as the following (see Figure 3):

• Challenges in identifying and using appropriate metrics. Historically, very few orgs have attempted to track metrics for DEIB and even fewer have ventured beyond collecting diversity data. Often, leaders are unsure which metrics can and should be measured for DEIB. Even if they’re able to identify them, leaders then often face challenges around tracking and integrating the data.
• Legal, security, and privacy issues. DEIB data involves sensitive information—and this comes with legal and security challenges around data collection, storage, and usage. As a result, some orgs hesitate to collect and use it. Additionally, employees may be hesitant to provide it, due to data privacy and access concerns.
• Poor alignment with goals. Orgs find it challenging to use the data if there’s no or poor alignment between the data collected and the overall DEIB goals that the company wants to achieve. As result, there can be a sense of helplessness, which can render the data not as helpful.
• Data responsibility issues. Because DEIB data can reside in multiple systems under several functions (e.g., HR, D&I, IT, sales), there can be a lack of clarity around who is primarily responsible for the data and how / when it can be shared.
• Data interoperability issues. Related to the previous point, orgs often find it challenging to use data collected in one system on another due to integration issues and capabilities of the tech solutions in place.

For this article, we focus on the first bullet to help orgs identify the range of metrics they can use.

“When you have members of a minority group who are leaving at a higher rate, that’s telling you something is wrong, and it helps steer you to where the problems are. It needs to be measured at quite a low level in the company because that’s the way you find where your hot spots are.”

—Fiona Vines, Head of Inclusion and Diversity and Workforce Transition, BHP⁸

Clarifying diversity metrics

As we highlight in our report “DEIB Analytics: Getting Started,” the essential first step to creating diversity metrics is collecting appropriate demographic data. Essentially, the data collected should allow orgs to answer 3 questions:

1. What does our current workforce look like across different levels (hierarchy) and functions / business units?
2. Who are we hiring (internally and externally) across different levels?
3. Who is leaving the org and at which level(s)?

It’s important that leaders not only look at simplistic diversity numbers, such as gender or race / ethnicity—they also need to consider multilevel diversity, known as intersectionality, such as Black women or gay Asian men. This additional analysis helps leaders understand their workforce at a more nuanced level, and make better recommendations and changes.

Many orgs track basic diversity numbers: 96% of U.S. companies report the gender representation of their employees at all levels and 90% report gender representation at senior levels.⁹ However, far fewer orgs look at intersectionality: Only 54% of companies track gender and race / ethnicity—such as Black or Latina women in senior leadership.¹⁰

Figure 4 is a list of common demographic data that we’ve seen orgs collect (for a more comprehensive list of data that could be collected, please see our definition in the earlier section). It’s important to note the significant legal limitations in different countries as to which of the following can be collected and stored. Your org’s legal counsel should always be involved in determining which data to collect.

While comparatively easy to collect and analyze, orgs should be wary of trying to do everything at once when it comes to diversity metrics. Leaders should first figure out the immediate challenges or business issues they want to solve for and identify the appropriate metrics accordingly.

Examples of diversity metrics

Figure 5 offers a list of the metrics that orgs can use to measure diversity. Many orgs already collect most of these metrics through their human resource information system (HRIS) or applicant tracking systems (ATS). By adding a demographic lens to these metrics, orgs can quickly understand the state of diversity within the org.

Understanding equity metrics

Equity metrics can help orgs understand the effectiveness of their processes, and identify unfair or biased systems, practices, and policies. Research conducted in 2021 revealed that when employees are treated fairly, they’re:¹²

• 8 times more likely to look forward to going to work
• 3 times more likely to have pride in their work
• 4 times more likely to want to stay a long time at their company

Equity metrics can be measured from data collected via several sources, such as:

• Learning and development data
• Performance management data
• Payroll
• Employee engagement / experience data

Ensuring fairness in the distribution of resources, opportunities, and access can help leaders address existing systemic inequities within the orgs. The point to note here is that the distribution needs to be fair, not equal. The difference between these two concepts is shown in Figure 6.

Thus, the goals of measuring and tracking these metrics should not be to ensure equality or sameness for everyone, but rather to:

1. Detect areas in which systemic inequities exist
2. Identify differences in capabilities, resources, and needs
3. Implement systems and process that take these into account

While orgs have a strong case for creating a fair and equitable environment, many struggle to do so. For example, our 2021 study on performance management trends revealed that only 48% of employees believe their performance evaluation process is fair and consistent.¹³ As orgs continue to manage unique needs and challenges for different employees, leaders will increasingly need to address issues around managing fairness and equity across varied employee experiences.

Source: Robert Wood Johnson Foundation, 2017¹⁴

Examples of equity metrics

Below is a list of metrics that orgs can use to understand, measure, and track equity. All metrics should be analyzed by the different demographics collected by the org to understand the differences in opportunities, access, and renumeration for various groups.

Identifying inclusion metrics

After diversity, inclusion is the most common area that organizations tend to measure. According to a 2018 study, a little more than 50% of orgs measured inclusion.¹⁷ While the focus and urgency around this area has increased over the years, few orgs are doing anything beyond tick-the-box exercises.¹⁸

“Let's say that the engagement score for our company is high at 80%, and that makes us happy. And then you realize that 80% of your employees are White—which means that you’re not really hearing the voice of those under-represented groups. Inclusion analytics is about pulling that out, and making sure you have a good sense of where everybody's falling on all of your core metrics.”

—Hallie Bregman, PhD, Global Talent Strategy and Analytics Leader¹⁹

There are a few reasons why orgs should focus on understanding and measuring inclusion. Orgs with an inclusive culture:²⁰

• Are twice as likely to indicate they met business goals from last 3 years
• Are 81% more likely to indicate high customer satisfaction
• Have employees that are 45% more likely to stay
• Have employees that are 2 times more likely to give a positive Net Promoter Score® (NPS)

If these reasons weren’t enough, the volatility of 2020 and 2021 has resulted in many companies facing tough questions around their efforts in this area. According to a recent analysis of S&P 500 earnings calls, the frequency with which CEOs talk about issues of equity, fairness, and inclusion on these calls has increased by 658% since 2018.²¹

Inclusion metrics can help orgs understand whether employees feel:

• Accepted by others in the workplace
• Integrated into and a part of the wider organization
• Respected for their work by others

As alluded to above, orgs can typically approach inclusion metrics in 2 ways—employee perception data and object data. We explain the differences between the 2 in Figure 8.

Examples of inclusion metrics

Figure 9 offers a list of metrics that orgs can use to understand, measure, and track inclusion. These include metrics that directly impact an employee’s sense of inclusion (e.g., mentor relationships and strength of connections with others), as well as some not-so-obvious metrics that can drive inclusion (such as the average distance between office and home, which can adversely affect employee experience).

Defining belonging metrics

While closely related to inclusion conceptually, it’s important that orgs pay equal attention to measuring and understanding belonging. We explain how belonging is different from inclusion in Figure 10. A high sense of belonging among employees can result in:

• An increase in employee happiness and employee engagement, which in turn impacts employee retention²²
• A significant increase in job performance²³
• A reduced turnover risk and a decrease in employee sick days²⁴

Analytics based on belonging metrics can serve as a leading indicator of critical diversity outcomes as well. Specifically, belonging metrics can help orgs to:

• Gain a deeper understanding of the sense of security experienced by employees
• Find out if employees feel connected with the org’s values and purpose
• Bolster their ongoing efforts around inclusion and equity

“When someone is experiencing a sense of Belonging, they feel freer, they feel more creative and their opportunity to potentially have an impact at work is significantly increased.”

—Kate Shaw, Director of Learning, Airbnb²⁵

Examples of belonging metrics

Figure 11 offers a list of metrics that orgs can use to understand, measure, and track belonging. While some metrics speak to belonging directly (e.g., a belonging index as part of an engagement survey), others should be used in combination with one or more additional metrics to gain a better understanding. For example, by looking at metrics around the number of resources groups offer and the participation rates for them, orgs can try to understand if employees feel supported. Employee feedback comments specific to these topics can provide even more context of the underlying issues.

DEIB metrics: Strengths & limitations

DEIB metrics are most effective when multiple types of metrics are combined to gain a clearer picture of DEIB holistically. (See Figure 12.) For example, by combining inclusion metrics with equity metrics, orgs can understand not only that different groups may be feeling less included, but also the specific reasons (e.g., unequal development opportunities or biased performance reviews) for it.

Using data sources for DEIB

Now that we’ve covered the specific metrics, let’s look at the data sources orgs can use for them. Orgs should keep a few things in mind when using such data:

1. All data should be looked at with a demographic lens. For example, the number of trainings accessed by the workforce would mean little unless analyzed to see if white women access training more often than Black women.
2. Data are more powerful when combined with other data. For example, data from the HRIS that shows exit rates should be combined with data from exit interviews, surveys, and employee comments on external review websites.
3. Connectivity between data sources is essential to being able to use the data effectively. Data interoperability, or the ability for different data between systems to work together, is a necessity in order for orgs to drive DEIB. As such, they should look for tech and tools that enable them to do that.
4. The partnership between DEIB and people analytics functions is critical. As we mention in our report “DEIB Analytics: Getting Started,” DEIB and PA leaders often come from different backgrounds and parts of the org, which mean partnership challenges may exist that must be addressed. The insights and expertise of both groups are necessary to use and interpret DEIB metrics effectively.

Common data sources for DEIB

Figure 13 shows that most of the data sources can be used for more than one DEIB area.

Beginning the DEIB metrics journey

Orgs at the beginning of their DEIB journey should try to answer the question: What’s the current state of DEIB within the org? As such they should focus on 2 things:

1. Understanding the state of diversity
2. Identifying “low-hanging” challenges—areas that need attention and are easy to quickly start working on

When it comes to selecting metrics, orgs should start with the basics, like:

• Getting their basic demographic data in order
• Measuring metrics around headcount, retention, and turnover to understand diversity
• Leveraging employee perception data—such as engagement surveys, feedback, and focus groups—to understand how different groups perceive DEIB at the org

Orgs should ensure that the selected metrics are clearly tied to overall strategy and that processes exist to track their progress.

A people analytics leader we spoke to mentioned creating a Python script to pull different metrics that they’re already collecting around talent acquisition, internal mobility, performance, engagement, and exit rate to understand where the biggest gaps are between different employee groups. This allowed them to quickly identify areas with the biggest gaps, start working on them, and track progress over time.

“The DIB world is so enormous, and you could do a thousand things. It's hard to understand where to start and where to focus your efforts. We should be intentional about identifying our biggest gaps. Every company has some problems around DEIB, but we should work on finding where our biggest internal gap is and focusing on that first.”

—Head of People Analytics, a large technology company

Moving up to an intermediate level with DEIB metrics

Once the orgs have a clear sense of where they stand or the “what,” they need to understand the “why,” such as:

• Why do certain groups experience a low level of inclusion and belonging?
• Why are certain groups being promoted at lower rates than others?

Orgs can begin to supplement existing data to gain a deeper understanding of the systemic issues that impact DEIB. When it comes to metrics, orgs should look at data from existing systems:

• Learning & development data
• Performance management data
• Payroll data
• Wellbeing data
• Data from employee feedback comments

A technology provider shared an example of a customer project that conducted text analysis on data from employee feedback to understand why promotion rates for women were low in a company. The analysis revealed that the existing initiatives to drive promotions favored men and received positive feedback from them, as compared with women. Some of the concerns that surfaced included difficulties faced by women around childcare and the inflexibility around work schedules. The analysis of the data allowed the company to identify the systemic issues that were negatively impacting promotion rates for women and their overall DEIB efforts.

“Metrics are a way to communicate what’s important. Orgs should limit themselves to how many metrics they push. It’s like the weather, I don’t want a million different metrics to know if the weather is good of not. Orgs should figure out the goal (what is ‘good’ weather) and the metrics should help achieve that.”

—Dirk Jonker, Chief Executive Officer, Crunchr

Using a mature approach to DEIB metrics

The questions orgs should look to answer at this stage are:

• How can we address existing issues and drive our DEIB efforts effectively?
• How can we measure progress longitudinally?
• What creative analyses or approaches might help us answer questions we haven’t yet been able to answer?

When it comes to metrics and data, orgs should consider complementing existing data with:

• Network data
• Communication data from sources such as emails, calendars, meetings, etc.
• Workplace tech data from tools used by employees to get work done such as Zoom, SharePoint, Slack, Teams, and Asana
• Employee reviews and comments on external websites

Orgs should consider using advanced approaches to people analytics such as connecting text analytics with social network data. Text analysis can help orgs identify existing gaps in inclusion. Network analysis can help identify influencers. Orgs can relay feedback to influencers and leverage them to fill those gaps and drive greater efforts.

DEIB is a continuous effort rather than a “once-and-done” approach. Orgs should look externally to compare their performance to avoid becoming complacent in their efforts and update their goals regularly. Specifically, orgs should look at how other high-performing orgs that rank high on DEIB are performing, instead of industry or national averages.

“When it comes to selecting metrics, don’t go with the flow, and get something off the internet or another company. How you define metrics really matters, and orgs need to be intentional about what and how they measure them.”

—Lydia Wu, Head of Talent Analytics and Transformation, Panasonic North America

Conclusion

When it comes to DEIB, orgs need to do more than provide training and courses to employees. They need to think about and approach it in a holistic manner so that it’s built into the way the business is managed, instead of something that’s an afterthought or special.

To that end, orgs need to:

• Understand where they currently stand and how are they perceived by their employees. They should know what issues currently exist.
• Understand why those issues exist. Orgs need to find out the reasons why they are falling short in those areas.
• Identify what can they do to fix them. Orgs should plan their targeted initiatives and interventions in order to get the maximum value and results from their efforts.

In order to achieve that, companies need to apply a greater focus, and put more emphasize on using metrics and data than they currently do. As we’ve mentioned before, the growing demands from customers, investors, and employees around more action on DEIB is likely to keep increasing. Orgs stand to lose a lot more if they do nothing, not just in terms of lagging performance, engagement, and innovation—but also in future talent that’s going to place a lot more importance on these issues going forward.

It's time companies take their DEIB data seriously. Moving forward, we hope to see a greater acceptance of and creative thinking around how these data and metrics can be used to enable all people and do their best work.

Appendix

Below we share our own as well as indices used by other organizations to help understand their DEIB culture.

Figure 18: Gartner inclusion index | Source: Gartner.²⁷

Figure 19: University of California San Francisco’s Belonging Index | Source: University of California San Francisco.²⁸

1. Creating a DEIB culture: The skills every employee needs, RedThread Research / Stacia Garr & Priyanka Mehrotra, 2021, https://redthreadresearch.com/deib-skills-report/
2. People analytics tech 2020, RedThread Research / Stacia Garr & Pri-yanka Mehrotra, 2020, https://redthreadresearch.com/pat-2020/
3. DEIB analytics: A guide to why & how to get started, RedThread Research / Stacia Garr & Priyanka Mehrotra, 2021, https://redthreadresearch.com/deib-analytics-guide/
4. Diversity, equity, inclusion & belonging: Creating a holistic approach for 2021, RedThread Research / Stacia Garr & Priyanka Mehrotra, 2020, https://redthreadresearch.com/deib-2021/
5. DEIB tech: A market maturing to meet the moment, RedThread Research / Stacia Garr, 2021, https://redthreadresearch.com/deib-tech-2021/
6. “DEIB & analytics: What the literature says,” RedThread Research / Priyanka Mehrotra, 2021, https://redthreadresearch.com/deib-analytics-lit/
7. “Data and stories,” RedThread Research, Workplace stories season 2: Integrat-ing inclusion / Stacia Garr & Dani Johnson, 2021, https://redthreadresearch.com/integrating-inclusion-podcast-episode-6/
8. “Top Ways Companies Are Measuring Their Diversity and Inclusion Progress,” LinkedIn / Bruce M. Anderson, November 11, 2020, https://www.linkedin.com/business/talent/blog/talent-acquisition/ways-companies-measure-diversity-and-inclusion-progress
9. Women in the Workplace, LeanIn.org, 2021, https://leanin.org/women-in-the-workplace-2021?gclid=Cj0KCQiAnaeNBhCUARIsABEee8U2f05l-yQCPjifXrdEA6786L5SCNL8k6Twoo9Zbpk2F6hPEfOY0noaAhnQEALw_wcB
10. Ibid.
11. “Industrial manufacturer increased diversity in hiring process,” Visier, https://www.visier.com/customer-stories/industrial-manufacturer-improving-diversity-inclusion/
12. “Why Is Diversity & Inclusion in the Workplace Important?” Great-placetowork.com / Matt Bush, April 13, 2021, https://www.greatplacetowork.com/resources/blog/why-is-diversity-inclusion-in-the-workplace-important#:~:text=Diversity%20and%20inclusion%20(D%26I)%20is,more%20commitment%20from%20their%20employees
13. Modern performance management trends, RedThread Research / Stacia Garr & Priyanka Mehrotra, 2021, https://redthreadresearch.com/performance-management-trends/
14. “Visualizing Health Equity: One Size Does Not Fit All Infographic”, Robert Wood Johnson Foundation, 2017, https://www.rwjf.org/en/library/infographics/visualizing-health-equity.html
15. 21 Ways to Empower Employees in 2021, Cultivate.com, 2021, https://cultivate.com/21-ways-to-empower-employees/
16. “How one company worked to root out bias from performance reviews,” Harvard Business Review / Joan C. Williams, Denise Lewin Loyd, Mikayla Bo-ginsky, and Frances Armas-Edwards, April 2021, https://hbr.org/2021/04/how-one-company-worked-to-root-out-bias-from-performance-reviews
17. D&I Metrics: Survey Analysis, Institute for Corporate Productivity (i4cp), 2019, https://www.i4cp.com/file/1985/download
18. “To Make Real Progress on D&I, Move Past Vanity Metrics,” Harvard Business Review / Cindi Howson, May 21, 2021, https://hbr.org/2021/05/to-make-real-progress-on-di-move-past-vanity-metrics
19. “Use both the data and the story,” RedThread Research, Workplace stories sea-son 2: Integrating inclusion / Dani Johnson & Stacia Garr, 2021, https://redthreadresearch.com/integrating-inclusion-podcast-episode-1/
20. Creating a DEIB culture: The skills every employee needs, RedThread Research / Stacia Garr & Priyanka Mehrotra, 2021, https://redthreadresearch.com/deib-skills-report/
21. “11 Trends that Will Shape Work in 2022 and Beyond,” Harvard Business Re-view / Brian Kropp & Emily Rose McRae, January 13, 2022, https://hbr.org/2022/01/11-trends-that-will-shape-work-in-2022-and-beyond
22. “Belonging Is Crucial: How Employees Are Feeling During the Pandemic,” Glint / Kris Kitto, JUNE 30, 2020, https://www.glintinc.com/blog/belonging-is-crucial-how-employees-are-feeling-during-the-pandemic/
23. The Value of Belonging at Work: Investing in Workplace Inclusion, Betterup.com / Dr. Gabriella Rosen Kellerman & Dr. Andrew Reece, https://grow.betterup.com/resources/the-value-of-belonging-at-work-the-business-case-for-investing-in-workplace-inclusion-event?camp=7012J000001YGmw&utm_medium=OAD&utm_source=HBR&utm_campaign=AMS19+-+Belonging+Report&utm_term=WP
24. Ibid.
25. “What belonging really needs to take hold,” RedThread Research, Workplace stories season 2: Integrating inclusion / Dani Johnson & Stacia Garr, 2021, https://redthreadresearch.com/integrating-inclusion-podcast-episode-5/
26. “Top Ways Companies Are Measuring Their Diversity and Inclusion Progress,” LinkedIn / Bruce M. Anderson, November 11, 2020, https://www.linkedin.com/business/talent/blog/talent-acquisition/ways-companies-measure-diversity-and-inclusion-progress
27. “How to Measure Inclusion in the Workplace”, Lauren Romansky, Mia Garrod, Katie Brown, and Kartik Deo, May 2021, Harvard Business Review, https://hbr.org/2021/05/how-to-measure-inclusion-in-the-workplace
28. “Resources for Increasing & Sustaining Belonging,” https://devlearning.ucsf.edu/belonging-index