When Equal Isn’t Equitable
Insights & best practices from Microsoft’s Inclusive Research team
By Kara Costa and Adoley Azu
Equality and equity are notions that are sometimes used interchangeably. Take for example, a situation where you have limited resources - dividing them equally is the fairest course of action, right?
But as those of us on Microsoft’s Inclusive Research team have learned, “fair” or “equal” does not often create equity.
When we started the Inclusive Research initiative in User Research, we operated from the assumption that, as product makers who aspire to empower every person on the planet to achieve more, we need to be great at connecting with and understanding the needs of ALL customers. So, our first idea was to start by including historically excluded customers.
However, as we learned more and dug into what that means and how we would do it, we came to realize that inclusion itself does not address the challenge. Because we’ve historically oversampled certain customers, we need to rebalance to get a true picture of user needs. So, we’ve adjusted our goal to better reflect the opportunity we have before us: to increase equity in our product making decisions by de-centering historically dominant populations in our research (who are typically oversampled) and elevating historically excluded voices to ensure their needs are heard and addressed.
Insights from Real People
The trick is…the system is built on inequity, so finding customers who have been historically excluded means that you can’t expect to find them in the system itself. We knew we had to disrupt the system of traditional customer feedback channels, so we built our own system by partnering with a nonprofit network!
To date, we’ve done 20 sessions with people who have unreliable internet, people in rural areas, racial and ethnic minority communities, people with disabilities, people who have lower incomes, people who don’t have a computer and only use their phone, LGBTQIA+ communities, people who don’t speak English, and more. It takes more time and money to include perspective from unheard voices, but this investment reflects our belief in inclusive design. Designing for one can extend to many, unlocking value for all users; consider the myriad of innovations designed originally for accessibility that we now can’t live without, like SMS (Short Message Service)!
The Digital Divide is real and we can make a difference
A lack of internet access equates to a lack of access to the kinds of opportunities that many of us take for granted (jobs, services, education, finances, etc.). Across all of our sessions with customers, certain themes have repeatedly risen to the top. Designing for these scenarios simultaneously bridges equity gaps for excluded customers and creates value for all customers who may, for example, temporarily lack bandwidth or only have their phone. Here are our primary insights:
There’s no substitute for customer connection
- Talk to customers to understand their needs and life experiences.
- Co-design solutions with excluded customers.
Screen size frames inclusion
- Design for smartphone dependent people (people who exclusively use their phone to access the internet).
- Create “Jobs to be Done” for smaller screens, especially mobile.
The majority of the world does not have high speed internet
- Design for lower bandwidths (for example 33% of Americans who have unstable internet access).
- Streamline load times and processing requirements for apps and programs.
- Build offline scenarios for productivity and cost-savings (some people pay by the minute for internet access, our products should not cost them more just to use them).
Design for accessibility and trust
- Design accessible language, reading, and learning level experiences without English language and/or Western bias.
- Create learnings and help-content that support all people.
- Address the distrust of tech by making first-class privacy experiences.
The Inclusive Research Playbook
Our Inclusive Research work means helping user researchers, designers, and product makers understand the importance of customer voice and equity within those customer voices. To that end, we developed an Inclusive Research playbook that positions research as a path to more inclusive product-making decisions and outlines best practices. We think of it as a living, ever-evolving document that provides researchers with an understanding of the key principles and essential behaviors of inclusive research, as well as a set of practical tools for recruiting historically excluded audiences. In the playbook, we identify three key behaviors along with supporting documentation for each:
- Identify, examine, and dismantle the impacts that your internalized oppression, biases, and privilege have on your work. This is about understanding and unpacking how we as researchers have been impacted by, participate in, and uphold systems of inequity and oppression — and it’s critical to mitigate the impacts of those on the research. Supporting materials are centered on the period before and during research, and include Self-reflection prompts, Learning series, Discussion series, and Microsoft Diversity and Inclusion training.
- Be aware and intentional about who is included, who is excluded, and the impact of that on your work. Supporting materials are intended to be used during protocol design and include questions and worksheets around: Who is centered? Who is not?, Target audience definition, Method and approach, Discussion guide and protocol.
- Elevate less privileged voices and needs to ensure they are heard and addressed by deliberately de-emphasizing historically dominant and privileged populations in our work. Right now on our journey, we recommend capping or limiting participation of historically dominant populations to be equal to or less than their representative percentage of the population (understanding that this is broad guidance and should be tailored based on the study objectives and its audience). Supporting materials are focused for use during execution and beyond, and include: Inclusive recruiting guidance, demographic questionnaire, and suggested caps on historically over-represented groups, Execution best practices, Interviewing guidance, Sensemaking, and storytelling.
The biggest thing we’ve learned is that there is so much we don’t know and so much more we want to learn. The next things we’re thinking about include how we can reinvent product-making systems and operationalize inclusivity and equity, how we can extend our work globally, how we can measure impact and activate others, and how we can integrate Lilly Zheng’s perspective on DE&I business cases. All told, we’re on a long journey and this is just what we’ve learned so far. Does any of this resonate with you? What challenges have you faced? What have you tried? We’d love to have a conversation with you in the chat. For the systemic change we’d like to see, we have a hunch it will take all of us coming together to share the work. How can we work with you?
The header image is by Tendril and all proceeding visuals are by Dave Savage and Datalands.
To stay in the know with Microsoft Design, follow us on Twitter and Instagram, or join our Windows or Office Insider program. And if you are interested in working with us at Microsoft, head over to aka.ms/DesignCareers.
