23andME

Inclusive Design

PROJECT DESCRIPTION

DNA Relatives is a 23andMe feature that predicts genetic relatives based on the percentage of DNA shared between two users. The current experience was not gender inclusive as it forced trans and nonbinary users to share potentially sensitive information regarding their genetic sex.

The goal of this project was to make sharing information safer and easier for intersex, transgender, and nonbinary individuals in addition to improving the overall user interface of the page.

CLIENT

23andMe

ROLE

Product designer

PROJECT TYPE

Exploratory design

🏢 BUSINESS GOALS

  • More inclusive design
  • More personalized experience
  • Increased user engagement

📈 SUCCESS METRICS

  • Number of opt-ins
  • Completion of profiles
  • Overall engagement

👵🏽 CUSTOMER GOALS

  • Safely share private information
  • Informed consent
  • More robust DNA Relatives experience

DNA RELATIVES PREFERENCES

Currently the DNA Relatives Profile page consists of three parts.

1. Basic Information - This section contains the user’s basic information such as profile name, sex, and birth year. Most of this information is prefilled from the user’s registration

2. DNA Segments - This section allows users to share their shared DNA segments and ancestry reports with other users. Some of this information may reveal sensitive health information.

3. Enhanced Profile  - Users can upload a profile picture and provide more information regarding their family history such as surnames and locations.

RESEARCH GOALS

I had two objectives in mind when conducting research for this project.

First, I wanted to understand the challenges nonbinary and transgender individuals faced when completing online profiles. This would help create an experience that they would feel safe and comfortable participating in.

Second, I wanted to identify and remove any pain points of friction within the existing flow.

INTERVIEWS

“When I see myself represented it gives me a lot of joy and confidence with the product. It goes a very long way towards me being excited as a customer and recommending it to others.”
To better understand the experience of transgender and non-binary individuals, I interviewed three individuals within the community. From these interviews I gathered the following insights:

USABILITY TESTING

In addition to making this DNA Relatives Preferences page more inclusive, I wanted to identify other ares of improvement that would make the page more usable for all our users. I conducted Usability Tests over zoom with three individuals in their early 30s.

COMPETETIVE ANALYSIS

While speaking with the nonbinary and transgender individuals that I interviewed, they mentioned several services that handled gender inclusivity well. I reviewed these websites to identify patterns that made for good inclusive design. These websites included LesbiansWhoTech, Facebook, and OkCupid.

PERSONAS AND USER JOURNEY MAPS

With this in mind, I created Personas and User Journey Maps to better understanding the emotions and frustrations of different customers using DNA Relatives.

I focused specifically on two personas whose genotyped sex did not align with their gender identity (a trans man and a 60+ year old man whose Y chromosome has degraded). Additionally, I created a persona for an individual that I would consider to be a power user of the feature (an adoptee searching for her bio family).

What I observed from all three types of users is that they had a negative user experience when the service did not guide them through decisions that required a better understanding of DNA than what they had. For example, Fred (a trans man) could have a negative user experience if he chose to share the "DNA Segments" portion of his profile if he did not understand that other customers might be able to infer that he did not have a Y chromosome from this information.

PROBLEM STATEMENT

SCENARIO

People would like to participate in DNA Relatives to engage with potential new matches.

PROBLEM

Transgender and nonbinary individuals may not feel comfortable or safe with how some of their information is currently shared to others.

GOALS

Provide a more modular approach to sharing information so users only have to share what they are comfortable with.

PRODUCT IMPROVEMENTS

  • Allow users to select pronouns for predicted relationships
  • Provide more context and flexibility around “DNA Segments” section of opt-in
  • Make profiles easier to fill out with recommendations for surnames and locations

DESIGN

WIREFRAMES

I then started creating wireframes to begin my redesign of the DNA Relatives preferences page. From these wireframes, I encountered two problems.

First, some of the users tested mentioned that the page was too long. For this reason, I broke it up into three separate pages.

Second, I realized that there were no gender neutral terms for some relationships (e.g. aunt, uncle, niece, nephew). I went back and consulted one of the participants that I had interviewed earlier. We ended up coming up with terms such as "parent's sibling" and "sibling's child". What we realized was that there were not always simple solutions to these types of problems.

GENDER FIELDS

Based on my user testing, it became apparent that I had to also revision how 23andMe was collecting data around gender and sex. Many nonbinary users had commented on how their gender and sexual identity was integral to their health care experiences. Ignoring the nuance in gender and sexual identity resulted in poor health care experiences as it ignored questions that were integral to these user's mental and physical well-being.

These are the following changes that I made to how this information would be collected:

HIGH FIDELITY DESIGNS

Based on my research, I created wireframes and high fidelity designs that I used for usability testing. The usability testing for the designs was overall positive especially in regards to the updates for improved gender inclusivity.

DESIGN SOLUTIONS

Field for Pronouns

Users would be able to select their pronouns and see how this information would be displayed to others.
Description of Shared Reports

Provide users more context around the DNA reports they would be sharing. User modulation tools would allow users to only share information that they are comfortable with.
Comprehensive Disclaimers
‍‍
Users would be notified if they were sharing potentially sensitive information. For example, if their genetic sex did not match their selected pronouns.
Recommended Surnames and Locations
‍‍
Recommending surnames and locations based on match data would make completing profiles quicker and easier.

CONCLUSION

Overall, the project was a success. Most of the users tested said they liked the flow and clarity of the newly designed pages. Additionally, nonbinary and transgender people interviewed stated that the updates allowed them to make more informed decisions which made them more comfortable participating in the feature.

RETROSPECTIVE

This project was a great learning experience on how to design inclusivily for all types of people. Speaking with nonbinary and transgender individuals allowed me to better empathize with this community and to understand the importance of designing for all different demographics. This was a great example of how designing for all demographics generally improves design overall.

Working with my design mentor at 23andMe in addition to my DesignLab mentor allowed me to address real world problems and to gather feedback from actual stakeholders.