Organisations are becoming increasingly focused on improving Diversity and Inclusion (D&I) in the workplace. Consequently, it’s imperative to understand the data behind these initiatives.
Diversity, Inclusion, and Equality are not easily achieved simply through increased spending. The lack of success of many D&I initiatives in recent years, such as how Google’s tech workforce was only 2% African American in 2019, despite reportedly spending $150 million in 2015 on diversity programs, is just one of many examples of this. Despite the efforts that many organisations are making to improve D&I, most have a long way to go. This is partly due to how we define and measure Diversity and Inclusion. These initiatives have traditionally been limited by the type of data an organisation has access to, and the types of analysis they are capable of performing. Due to technical advances and new methodologies, it is now possible to understand and improve both Diversity and Inclusion through data analytics. Before describing how this can be done, let’s start by making a clear distinction between the two:
Diversity refers to the traits and characteristics that make people unique. In the context of an organisation, this is the mix of people and demographics within the workforce, and in different areas and levels of the organisation.
Inclusion refers to the behaviours and social norms that make people feel welcome. In the context of an organisation, this is the feeling that all people have equal access to the same opportunities.
The key to a successful D&I program, is to improve both of these aspects. The saying “What gets measured, gets managed”, remains relevant here… but how do we measure how diverse or inclusive an organisation is?
Arguably, most D&I strategies focus almost entirely on Diversity. This is not only because it is easier to measure, but also because it is easier to manage. Measuring Diversity usually involves analysing what percentage of the workforce falls into a certain demographic. This includes looking at specific areas of the organisation or tracking the career paths of members of each group.
Taking gender diversity as an example. Organisations would typically measure the ratio of men to women within each team or department, and among new hires. They would also consider looking at differences in pay, opportunities for promotion, and representation in senior roles. Diversity can be influenced directly – for example with affirmative action initiatives. This guarantees results (often short-term), by hiring or promoting members of the under-represented group. Unfortunately, many companies stop there… at least from a data perspective. While Diversity can be the easier target when it comes to measuring and managing the data, it is difficult to produce lasting, long term effects without also tackling Inclusion.
Inclusion is of course a far less tangible characteristic of a workforce. It is deeply embedded in social norms, company culture, and unconscious biases that are very difficult to describe, let alone measure. It is often targeted through initiatives such as training and seminars, appointing Inclusion champions and sponsors, or even just hiring for Diversity and hoping that the company culture will change with it. These initiatives are certainly helpful, and should be part of any D&I strategy, but without capturing the right data, how do we know if they are even working?
Organisational Network Analytics (ONA) is the process of making the invisible lines of communication within the workforce visible. It is a method of capturing and analysing the formal and informal relationships between the people within an organisation. Using ONA, it is possible to capture and track social data that can shed light on how inclusive a workforce is. There are two main methods of capturing this type of data. The passive approach involves monitoring the communications (the people involved, not the contents of messages) between employees on platforms like Outlook and Teams. The active approach, which was used in the example below, involves surveying employees. These surveys would include questions such as ‘Who do you depend on for knowledge?’. This type of data can be used to uncover the true dependencies and connections that exist between members of the workforce.
Returning to the gender diversity example. Instead of looking at how many men and women make of up the workforce, we can start to analyse why men are more likely to get promoted than women in many sectors. This can be done by understanding what types of social networks each group has, and looking for differences between the average male profile compared to the average female profile. In one study, it was found that men are more likely to have a smaller number of unique connections, with mostly people who are more senior than they are. While on the other hand, women were more likely to have a larger number of unique connections, but with people of the same or lower grade as them. However, among both groups, those who were promoted had more frequent access to senior managers. This kind of insight into the behaviours of a workforce, can be used to measure and influence Inclusion. It allows us to better understand what aspects of the culture should be targeted, and to monitor whether that culture is changing as a result of D&I initiatives.
To learn more about ONA, you can read our recently published whitepaper – it has information about how to get started, from practical tips and examples to learnings from the field.
Also, watch this space – I plan to produce a follow-up blog where I bring to life, with a dashboard, how ONA can help with Diversity & Inclusion.
At TrueCue, we have People Analytics and Organisational Network Analytics experts, who have built Diversity and Inclusion solutions that will help you track and manage your D&I programs.
Mark is a Data Scientist at TrueCue. His equal interest in business and technology, along with his belief that they are better together, is the result of an academic background in Business Studies, Computer Science and Data Science.
This belief, coupled with his proficiency in many programming languages and technologies including Python, R, SQL, PowerBI, and Tableau, allows him to solve business problems by pushing the correct tools to their limits. He has applied this mix of skills to his current area of expertise, Organisational Network Analytics.
When he isn't working with data, he can usually be found walking his three dogs, or climbing up the face of a large rock.