We all know that diversity, equity, inclusion, and belonging (DEIB) are more important than ever. The last 18 months—between COVID-19 and the social justice movements—have heightened our awareness of how much work is yet to be done when it comes to DEIB.
As a result of this experience, the expectations of 3 important groups of stakeholders have changed:
- Customers. Consumers now expect organizations to get their DEIB house in order. For example, 80% of consumers want brands to help solve society’s problems, and 64% want companies to set an example of diversity within their orgs.
- Employees. Likewise, employees also expect their employers to do more. According to the Edelman Trust Barometer, in 2021, 50% of employees said it’s more important than last year that an employer have a diverse, representative workforce.
- Investors. The SEC now requires public companies to disclose information about their human capital practices, which can include their DEIB metrics. Due to the uneven reporting, however, there’s talk about further reporting requirements coming out soon.
A significant increase in focus on DEIB has come with this change in expectations. However, as we all know, we need data to measure change, so there’s much greater interest in DEIB analytics now.
Historically, organizations have collected diversity data to fulfill legal requirements (the Equal Employment Opportunity Commission in the US, for example). As a result, organizations have collected at least basic diversity data for decades. For example, a 2019 survey found that 92% of organizations measured workforce diversity (or planned to within the year).
By contrast, inclusion is less frequently measured: That same 2019 survey found that just 77% of organizations measured inclusion (or planned to within the year). Unfortunately, we don’t know the breakdown between the percentage of companies measuring it and those planning to measure it: But we’d guess the actual percentage of companies measuring it isn’t more than 50%.
Yet, measuring inclusion is important for at least 3 reasons:
- People. Inclusion data allows the organization to understand what’s happening with its people (i.e., their connection to others, their sense of being valued) and how that may vary by diversity characteristics.
- Process effectiveness. Through inclusion analysis, leaders can identify and then address unfair or biased systems, practices, and policies.
- Prediction. Inclusion analysis can identify factors that may lead to changes in diversity data, such as decreasing employee engagement rates, or changing promotion or hiring rates among specific demographics.
Given the need for inclusion metrics and analysis—and their relative low level of use—this article provides the foundational information leaders need to get started on inclusion analytics. It’s based on our report, DEIB Analytics: A Guide to Why & How to Get Started, which is the result of a dedicated 6-month study of this topic added to our almost decade of researching the DEIB and analytics spaces.
5 Things to Get You Started with Inclusion Analytics
There are at least 5 things you need to know to get started on inclusion analytics:
- A clear definition of inclusion analytics is important
- You have to get your diversity demographic house in order
- Consider starting with employee engagement data
- You have to find critical problems people care about
- It’s important to share information broadly
We’ve detailed these 5 keys to getting started in the following sections.
A Clear Definition of Inclusion Analytics Is Important
It’s always important to begin with a solid definition—and it’s especially the case here. We offer our definition:
Inclusion metrics and analytics are based on employee perception and objective data to understand the current state of inclusion in an organization.
Typically, organizations can approach inclusion analytics in 2 ways:
- By analyzing employee perceptions of their personal levels of inclusion within the organization
- In analyzing objective inclusion data by diversity characteristics to understand different experiences
Figure 1 shows several examples from leaders on the types of analyses that are typically done to understand inclusion.
Figure 1: Examples of Data & Analyses to Understand Inclusion
Source: RedThread Research, 2021.
Every organization’s likely to have different ways of thinking about and measuring inclusion. The most important point is that your organization defines inclusion for itself—and then uses that definition consistently.
You Have to Get Your Diversity Demographic House in Order
As leaders start on the DEIB analytics journey, they often feel overwhelmed by the number of moving pieces and can be, therefore, uncertain where to start. Given this, we want to be clear: Get your diversity demographic house in order before you measure inclusion. Without a solid understanding of who’s in the organization from a demographic perspective, you can’t understand how inclusion varies.
The most typical diversity data that organizations track include the following:
- Age / generation
In the United States, organizations also often track:
- US race / ethnicity
- Veteran status
- LGBTQ+ status
This information is then used to analyze inclusion data, such as that identified in Figure 1.
Consider Starting with Employee Engagement Data
We suggest organizations start their inclusion analytics journey by analyzing their existing employee engagement survey data. Many organizations already have some type of employee engagement survey in place: Often those data are already mapped to individual employees or to their diversity characteristics.
After examining their existing engagement survey instrument, many leaders identify a subset of questions that can be combined into an inclusion or belonging index. Other organizations incorporate new questions into their existing employee engagement surveys—or they create an entirely separate inclusion or belonging index. Regardless, it’s important for organizations to identify a primary mechanism to measure inclusion.
For our research, we have developed a DEIB Index (see Figure 2), which could serve as inspiration or a model for your own use.
Figure 2: RedThread’s DEIB Index
Source: RedThread Research, 2021.
Leaders then analyze the engagement or inclusion / belonging data by diversity characteristics. It’s extremely important to remember to look at intersectionality—when more than one diversity characteristic is present (i.e., Black women or gay Asian men)—in doing this analysis, as long as the data set is big enough to provide adequately large subpopulations. Looking at just one characteristic can obscure what’s really happening.
It’s extremely critical to involve the team typically responsible for the overall employee engagement analysis—such as the people analytics team—in this work. That team can help identify true anomalies versus overall patterns in the data. For example, it may seem like Latinas are not very engaged in one function. However, a comparison to the overall data set may show that Latinas have the same engagement rate as employees in the broader company, while the rest of the organization has an unusually high engagement rate—thus making the Latina engagement rate seem low.
The point of this work is to establish a solid baseline, which can be used to understand the current state (and most pressing problems)—and be leveraged in the future to measure change. Stakeholders must be aligned and accepting that this measurement is the baseline. Many organizations integrate these inclusion data into other management dashboards, so that leaders can regularly see them in the context of other management data.
You Have to Find Critical Problems People Care About
Armed with the anomalies in the inclusion data, it’s important to then identify the most important problems to solve. There are many different ways that leaders prioritize problems. Here, we include some of the questions our interviewees have used to help them prioritize the inclusion analytics problems they want to solve:
- To what extent is this problem aligned to the big-picture DEIB strategy and priorities?
- What will be the top- or bottom-line impact?
- How many employees will it impact?
- Is this problem a recurring, long-term one or an acute need?
- What will be the level of effort and resources required to understand the problem?
- Will we be able to yield actionable insights that can result in behavior change as a result of this analysis?
- Who is asking the question? (CEO, CFO, CHRO, etc.)
The point of this prioritization exercise is to ensure that efforts spent on inclusion analytics lead to change instead of the analysis sitting on the metaphorical shelf.
Once the challenges are prioritized, leaders then conduct additional analyses to understand what could be driving low levels of inclusion in a given population. For example, let’s say that Asian men’s sense of inclusion is low and the specific problem is that they don’t feel there’s equitable access to opportunities. Leaders could then conduct additional analyses to see if that population is accessing development opportunities or being promoted at the same rate as the majority population. Leaders could also analyze if something is happening with this population’s tenure in role, tenure at the organization, or their performance scores.
As you can see, this is an iterative process to understand what’s happening with specific groups when it comes to inclusion. But, because of the prioritization process, leaders can take comfort in knowing that their hard work should lead to insights which then result in a decision.
It’s Important to Share Information Broadly
For organizations that are new to the inclusion analytics journey, it may feel like inclusion metrics and analytics should be held close to the vest. When it comes to any personally identifiable information, that inclination is correct. However, for de-identified information, it’s important to share relevant inclusion metrics broadly. If managers can’t see inclusion metrics for their teams, then they can’t make progress and feel ownership for them.
Generally speaking, organizations are still early on their journeys of sharing metrics. As shown in Figure 3, companies tend to share inclusion metrics with senior leaders, but much less so with employees. That said, we’re beginning to see companies increasingly share real-time inclusion data with employees, similarly to how they’re increasingly sharing engagement data. This requires those organizations to work through data privacy and security protocols, but they are finding ways to do this effectively. This increase in transparency increasingly enables managers at all levels to understand what drives inclusion and to take steps to improve it.
Figure 3: Frequency of Organizations Sharing Inclusion Metrics
Source: i4CP, 2019.
We hope this article gives you a starting point for understanding inclusion analytics and how you can begin your journey. This article represents just a snapshot into many of the insights we’ve learned from this research. As part of this topic, we have a full study, DEIB Analytics: A Guide to Why & How to Get Started, as well as a free infographic that summarizes some of our findings. Please get in touch with us at email@example.com if you’d like to learn more about our insights on DEIB analytics.
Diversity Analytics: A Guide to Why & How to Get Started, RedThread Research / Stacia Garr & Priyanka Mehrotra, August 2021, https://redthreadresearch.com/deib-analytics-guide/
DEIB Analytics: The 8 Steps to Get Started, RedThread Research / Priyanka Mehrotra, September 2021, https://redthreadresearch.com/deib-analytics-infographic/