Inclusion Survey.

Our updated process

In an effort to bring an inclusion and diversity lens to everything we do, we have shifted to make deeper inclusion analysis a standard part of our engagement survey process, rather than a stand-alone survey. You can find our updated process here.

Asking why and how

This post is about how we run our annual Inclusion Survey (in a separate post, we’ll discuss the actual results of the survey). We find the survey to be powerful in assessing inclusion across a broader spectrum of diversity. We’re open-sourcing questions from our Inclusion Survey so you can run one at your company. Download the PDF of the questions we used.

Why we do this

Diversity is about much more than the demographics (race/ethnicity, gender, and age) we, as a company, have to report under federal law. We believe government-mandated forms oversimplify the broad range of identities that we’re trying to create more space for every day. So we run a voluntary, anonymous Inclusion Survey to give our employees a way to be measured as they self-identify, and to give us a way to measure whether we’re truly building an inclusive workplace where people from all backgrounds can thrive.

The survey measures self-reported demographic information in categories beyond race and gender, including:

• Ability
• Age
• Caregiver status
• Education
• Education level of parents/guardians
• English proficiency
• Family status
• Gender identity
• Geographic origin
• Immigration generation
• Political views
• Race/ethnicity
• Religious views
• Sexual orientation
• Socioeconomic background
• Transgender
• Veteran status

We also ask employees to rate their experience in categories including:

• Fairness
• Belonging
• Decision-Making
• Voice
• Opportunities and resources
• Personal importance (i.e. the level to which one cares about I&D)
• Taking action

Part of creating a culture where we’re celebrating identities means allowing people to name them! We hear that the survey itself helps people feel seen. And by evaluating employees’ sense of inclusion alongside self-reported demographic information, we can identify areas where we’re falling short and develop plans for improvement. This isn’t about hoping for rah-rah results so we can pat ourselves on the back about how great we are at I&D. We want to build a survey that gives us signals on where to improve. So we dig into sensitive areas with the understanding that results will be mixed.

How we do it

This was our second year of surveying. In 2017, we partnered with CultureAmp to run our first survey, and many of our questions about inclusion are adapted from theirs. We’re grateful that they’re encouraging more companies to measure inclusion health by open-sourcing their questions—you can read more about that here.

We give employees one week to complete the survey, and we send reminders over several different internal communication channels. A big part of driving participation is helping folks understand how we’ll use this data. We reinforce that it will drive our external report-outs, and more importantly, inform actions we take over the next year. As the week goes on, we check that our participation is representative by comparing the breakdown of self-reported demographics from the survey to the demographics we have data on in our HRIS system (like org, location, gender, and race/ethnicity). We also explicitly remind folks that it’s anonymous. To that end, responses aren’t tied to employee identifiers like name or email, all questions on the survey included a “prefer not to say” option.

What we’ll do next

As you can imagine, we have a lot of data. It’s amazing.

Externally, in an upcoming post, we’re going to dig into our demographics and the insights we’ve gained as part of our commitment to sharing data publicly. Internally, we’ve already shared the overall demographics and insights within Square to hold ourselves accountable and create conversation, because that’s what we’ve found to be most successful in driving progress.

Most importantly, we are using the data to drive specific actions. We look at both individual demographics and intersectional views of employee sentiment. Combining the expanded demographics with the inclusion questions means we can be extremely specific in how we solve challenges. For example, what happens if we see that single parents in their 30s are scoring lower on questions of voice and resources? We can identify actionable next steps from these insights: let’s offer more programs for single parents. There’s more work to do gleaning intersectional insights, and we’ll keep mining the data to inform new People programming as we look to 2019.

The survey is still a work in progress. Feedback indicates we have a few areas where we can improve, particularly globalization. We use the survey worldwide, and global offices reported that some questions could be improved to be less US-centric—for example, race identification isn’t consistent from country to country. We also need to be mindful of word choice when using words like “immigrant” or “expat,” similar to how we make an effort to use the word “global” instead of “international.” This year, 95% of respondents found the survey inclusive, up from 86% last year, and we’re working to close that last gap.

We’re excited to share some of our learnings in our next post. In the meantime, we’d love to know how you’re measuring your inclusion health and how you’re using those insights.