How we set diversity targets
This post is part of our D&I series. For more information, please take a look at the Versett 2017 diversity report.
When designed thoughtfully, Diversity and Inclusivity (D&I) helps everyone. There is currently no standard D&I checklist or guide to best practices to follow — every company, department, or team will have different targets. In this blog post, we hope to provide a platform for learning, sharing knowledge, and providing useful resources that we found helpful at Versett.
The first step to achieving goals is setting them. Targets are achievable, time-based goals that help to focus attention, clarify accountabilities, and demonstrate a commitment to deliver. Setting realistic diversity targets based on analysis and baseline data increases the chances of achieving those targets.
Step 1: Survey
When setting targets, you need to know where you currently stand. An easy way to start is by sending out a company-wide survey to better understand organizational D&I health and areas for improvement.
Just like in user testing, good survey design is an art and a science. D&I surveys in particular tackle highly sensitive information. After conducting research on diversity surveys, our team created an inclusive survey to collect demographic information. We found many existing surveys lacking especially pertaining to subjects of gender, race, ethnicity, and disability.
This process was a huge learning experience for us — between team feedback and further research, our survey and survey questions are still changing and evolving. For example, we used the words “male” and “female” in the original survey when discussing gender. We’ve since debated whether or not to switch to “man” and “woman” because “male” and “female” connote a biological categories and can seem disrespectful.
Step 2: Identify the groups for which you will be setting targets
The next step is to consider the different identifiable groups for which you should be setting targets, e.g. managers, seniority levels, leadership, and departments. Different groups will have significantly different baselines as well as opportunities to achieve targets. Leadership and management of tech companies especially are overwhelmingly cisgender, White, and man; a target of 50% women might be a considerable challenge for leadership while readily achievable among other groups.
Versett is a small but growing company. For our first year of D&I, we’ve set targets for our organization as a whole instead of setting targets for different groups. Our smaller offices, leadership team, management team, and separate departments do not currently have a large enough number of people on their own for target setting. When they do, we will institute group-based targets because it is essential to push for accurate representation in all of these groups.
Step 3: Know the demographic you want to reflect
Ideally, the diversity of companies and their leadership would accurately reflect the demographic composition of the country they’re in; not just the state, city, or locale where they do business. Particularly in the United States, residential segregation of marginalized groups has been imposed by explicit federal, state and local policy as well as private acts of prejudice. This segregation impacts the demographic composition and racial and ethnic breakdowns of cities and neighborhoods. Setting diversity targets based on state, city, or locale demographics would invariably be affected by this discrimination. We therefore focus on the demographic composition of the country as a whole in order to set diversity targets less affected by this discrimination.
Large organizations like Google are expected to have workforces that accurately reflect the demographic composition of the country they’re in. With thousands of employees and over 20 offices in the US alone, they have seemingly endless resources and an extensive, diverse available talent pool to draw from. Even so, such organizations display a severe lack of representation of marginalized groups. Having accurate representation can be even more difficult for smaller companies that may not have the same level of resources and available talent pool.
Versett currently has offices in Calgary, Alberta and Toronto, Ontario in Canada as well as New York City, New York in the United States. New York City is one of the most diverse cities in the United States; Ontario is the most diverse province in Canada. In an effort to reflect the diversity and available talent pools we have access to we’ve proportionately drawn together the demographics of the Alberta, Ontario, and New York City populations (rather than that of the United States and Canada as a whole) to create our diversity targets.
Another approach we found is the idea that companies should consider reflecting the demographic composition of the client and user base they currently and/or desire to serve if it is more diverse than the demographic composition of the country they’re in. A workforce that reflects its diverse userbase can better serve, connect with, and understand the challenges of their users, increasing user satisfaction and the likelihood that your products will meet the needs of your diverse customer population. Relatively homogenous client and user bases should not be considered when creating diversity targets as their needs are better met by diverse and inclusive companies that drive innovation and create better products.
Most of our current and past clients are located in the states and provinces where we have offices. As our client and userbase expands, we will revisit our D&I targets.
The last approach we found is the idea that companies should reflect the demographic composition of the workforce of the country they do business in. The demographic composition of a society’s workforce usually differs from the overall population; for example, White people make up more than 80 percent of workers in the United States while the population is approximately 61 percent White, with Whites seeing higher wages and lower unemployment. Minorities have a harder time getting jobs and are paid less than similarly situated White workers.
We decided to base our diversity targets on the demographics of the overall working-age population in Alberta, Ontario, and New York City. We defined this as aged 18–67 to account for anyone from college students to the age of retirement in the US when you receive full benefits.
Step 4: Census information & number crunching
After selecting demographic information to use as reference, it was time to take a look at census data. It’s easy to forget you have this highly informative, publicly accessible resource at the tip of your fingers. While the websites leave a lot to be desired when it comes to visual representation and navigation, the United States Census Bureau QuickFacts provides statistics for all states and counties, and for cities and towns with a population of 5,000 or more. Requiring a little more number crunching, the Canadian Census Program provides vast sample data via Statistics Canada.
We also discovered that setting hiring targets may not be appropriate for all demographics. For example, because of stigma and methodological barriers it’s difficult to get an accurate count of the LGBTQIA+ population. Due to fear of homophobia, stigma, and discrimination, individuals tend to be hesitant to identify themselves as LGBTQIA+. Census data reports between 1% and 4% of the population identifying at LGBTQIA+. More secure polls have found 10%-19% of people identifying as LGBTQIA+. Research on younger generations have found even higher percentages of people identifying as LGBTQIA+. As a result, rather than setting hiring targets for the LGBTQIA+ community, we aim to institute employment protections and show that we are a safe and friendly work place.
Step 5: Be realistic
To set realistic, achievable targets you need to take a long-term approach and be pragmatic about possible growth, contraction, promotions, and restructuring. It’s difficult to achieve a significant shift in the numbers of diverse hires, especially in key and leadership positions, over a relatively short time-frame. It’s also difficult to set and achieve targets unless they’re based on realistic recruitment rates and internal restructuring.
Here’s the equation we used to set our long-term targets:
((New York City demographic percentage x Expected total number of NYC employees) + (Ontario demographic percentage x Expected total number of Toronto employees) + (Alberta demographic percentage x Expected total number of Calgary employees)) / (Expected total Number of employees) = (Long-term target demographic percentage)
We currently have a total staff of between 20 and 30 people; women comprise 27% of the entire group. Over the next year, we’re looking to hire ~10 people. With a negligible turnover rate (3% for both men and women over the past 3 years), hitting our long-term target of 50% women would require a recruiting 90% women (9 out of 10 recruits). Since this recruitment rate seems unachievable, we set the 2018 gender target at 37.5% women, requiring a recruitment rate of 60% (6 women from 10 recruits). We think this is an achievable target due to our current hiring pipeline and processes.
Setting realistic, achievable diversity targets is just one piece of the D&I puzzle. We’ll be reporting annually on our D&I progress as well as regularly sharing knowledge and strategies on Medium. As progress is made or unforeseen consequences arise, we’ll massage our targets to reflect realistic and ongoing goals. We’re excited to continue to challenge ourselves with this initiative.
NOTE: You may have noticed the captions beneath each illustration. We want to make image descriptions available to folks who are blind or have low vision. Medium does not currently support alt-text for images, but we’re hoping to encourage that to change.