UX Research for Safe Place International's TDAConnect platform

Safe Place International (SPI) is a leadership development organization for displaced LGBTQIA+ persons, mostly living in Africa. The Dream Academy (TDA) is their 10-week leadership development, socio-emotional learning, and employability skills course.

I worked on Phase 2 of the development of TDAConnect - an online community and resource center for TDA graduates.

The Team:
6 UX Researchers
4 Project Coordinators
13 UX Designers
6 Content Designers
6 Product Managers
6 Developers

My Role:
UX Researcher

Timeline & Methodology:
Phase 2 was completed in a highly collaborative agile environment over a total of 8 weeks, divided into four 2-week sprints. Research included both generative and evaluative methods including competitive analysis, user interviews, usability testing and card sorting.











We wanted to better understand how we might provide access to crucial resources and build a safe environment for TDA Graduates to connect, learn, and come together as a community.

Examining the Competitive Landscape

Since there was limited previous phase research pertaining to a Resource Center or Groups Feature, we began with competitive analysis to quickly identify opportunities for design teams before conducting user interviews.

Resource Center

Competitive Analysis & SWOT (with UXD2)

  • Competitors Analyzed: 7
  • Identified unique value propositions, advantages and disadvantages
  • Opportunities from SWOT prioritized with dot voting
  • Resource Center should be community driven
  • Resources should be organized and easy to filter based on need and location
  • Highlight sensitive information
  • Establish trust in external resources
Groups Feature

Heuristics Evaluation & SWOT (with UXD1)

  • Competitors Analyzed: 6
  • Established usability standards
  • Opportunities from SWOT prioritized with dot voting
  • Integrate code of conduct
  • Option to create private & public groups
  • Group Recommendations based on user interests, mutual friends, etc.
  • Ability to invite and share groups
  • Community-moderated groups

Learning from TDA Graduates

We leveraged assumptions based on previous phase research along with competitive analysis findings to inform user interview scripts and develop specific research goals for each feature. We then conducted two rounds of user interviews with 15 TDA Graduates to challenge assumptions and learn about needs, goals and current behaviors around finding resources and participating in online groups.


Interview participants were all located in Africa and the biggest challenges we encountered were data, power and connectivity issues, each of which presented significant scheduling difficulties. The circumstances necessitated a high level of flexibility and patience from both interviewers and participants.

Uncovering Common Themes

Resource Center
"I want to broaden my horizons. I want to learn more."


Participants are enthusiastic about contributing to TDA community andvoiced need for ongoing emotional support.


Participants used referrals to ensure legitimacy and protect against threats including scams and human trafficking.


Participants emphasized the importance of continued learning and access to mental health resources.

Groups Feature
"After graduation, maybe not just part. Maybe we keep the bond."


Above all, participants were overwhelmingly motivated by the desire to stay connected with one another post-graduation.


Participants voiced the importance of vulnerability and authenticity in connections and self-expression.


Participants continued to bring attention to concerns around online security as well as data, connection & power considerations.


Assessing Design Decisions

We collaborated with UXD to identify research goals and create a usability testing script for iterations on previous phase designs, including sign-up and onboarding flows, discourse feed, and adding friends. We then conducted moderated usability tests with 6 TDA Graduates to evaluate the working prototype.


Data, power and connectivity issues were specifically challenging to usability testing as participants' data limitations did now allow for them to click through the prototype on their own devices. Instead, we had to pivot to screen-sharing and asked participants to guide moderators through the prototype. Although this required adaptability, it was an incredibly valuable insight into very real pain points and challenges to consider in design.

Identifying Critical Issues (and wins)




Sign Up

  • Too much text in WhatsApp screen login / unclear CTA
  • Simplify text, allow users to copy access code and clarify instructions

Onboarding & Quick Exit

  • Onboarding flow too long. 66% did not read carefully
  • 50% skipped Quick Exit test
  • Shorten content to make it more digestible, use visuals to support
  • Introduce Quick Exit test outside of onboarding

Post Creation / Social Feed

  • Not straightforward how selecting a sensitive topic will affect their post
  • Indication of sensitive topic(s) chosen by user
  • Once posted - should be hidden to indicate that it included a sensitive topic

Friend Requests

  • Users assumed they were required to include a message to send friend request
  • Adjust hierarchy (progressive CTAs) to make clear that sending message is optional and not required

Refining Information Architecture

In Sprint 3, we used OptimalSort to conduct a moderated cord sorting exercise with 6 additional participants to deliver hierarchical and organizational insights to design and content teams for the Resource Center.

Data analytics generated by OptimalSort gave us our quantitative findings from the card sort while further questioning and observation provided deeper insights into why participants made their choices.


  • HIV Information, Gender Identity and Sexual Health  were split between Physical Health and Emotional & Mental Health
  • Recommendation to change category from Physical Health to Physical and Sexual Health
  • Discussion and findings validated foundational research suggesting the importance of emotional health


Strengthening Community

We learned that above all, community was the most evident overarching theme amongst the TDA Graduates that we spoke to in our research for the Resource Center. This finding informed client & design decisions across the platform.

  • Initially, the client intended to offer TDAConnect to all both graduates and current TDA students, but we were able to demonstrate through our research findings that expanding the user scope could detract from community trust within the platform. Ultimately, the client decided to limit access to TDA graduates only.
  • UXD incorporated the option to add a banner to profile pictures (example: indicating SPI Admin or members of the Wellness Team) to give users an additional way to contribute to their community.
  • Research uncovered that referrals in job searching both help to protect against potential threats as well as boost chances for employment, especially for undocumented refugees. With this in mind, UXD focused their design on a community-driven job board that builds trust through verification by highlighting trusted sources and emphasizes community impact.

Prioritizing Emotional Support

In user interviews, we learned that TDA Graduates face many challenges post-graduation and since they are displaced from their home countries, they often do not have access to a strong support system outside of SPI. There was a clear need amongst users for continued access to mental health resources as well as immediate emotional support and crisis response. To address this need, UXD incorporated a top-level banner in the Resource Center with quick links for immediate help, connecting users with SPI Wellness Team members.

Supporting Continued Learning

The TDA Graduates we spoke to were unified in their dedication to continuing their educational journeys. This insight sparked collaboration between UXD and the client to pivot from the initial plan of hosting the TDA curriculum on SPI's website to hosting the curriculum within TDAConnect, giving graduates quick and easily searchable access to past course materials as well as post-graduate continued learning resources.

Considering Low-Bandwidth & Data Limitations

In every user interaction, we were reminded of the barriers to connection that TDA Graduates face every day due to power rationing, data expense, and low-bandwidth limitations. These pain points shaped design and development decisions across the platform.

Design Considerations:

  • Limit number of posts or comments that can load at a time – use "load more" instead of infinite scroll
  • Minimize clicks necessary to accomplish a task

Dev Implementation:

  • Reduce size of client while maintaining functionality & ensure fast loading speeds on slower networks
  • Switched from React to Preact (faster, more efficient and lighter-weight option)
  • Switched from Redux to Zustand ( simpler more lightweight solution)
  • Switched from Bootstrap to Tailwind CSS (modular approach to styling to include only styles we need)


Connecting Users Through Groups

Since the majority of our user research for the Groups Feature was conducted and synthesized late in Phase 2, our insights and recommendations were recorded in handoff documentation for future phase design teams.

Design Recommendations for Future Phases:

  • Incorporate group space for users to share and collaborate on passion projects
  • Add of events pages to help facilitate group members connecting virtually and in-person
  • Create Admin page for group management & moderation to help create a safe experience

Addressing Privacy & Security

There are many TDA Graduates who are not able to be open about being a member of the LGBTQ+ community due to substantive threats to their safety. Insights from user interviews uncovered fear and distrust relating to data privacy and security. We posed this question to the development team: 

How can we reassure users that their data is secure and help them feel safe while engaging in the TDAConnect community?

Development Recommendations for Future Phases:


  • Encryption mechanisms must be implemented to protect sensitive data
  • Input validation libraries or frameworks must be used to ensure secure handling of user-supplied data


  • Collect minimum amount of data necessary for application to function effectively. Never store sensitive information on web client
  • Text inputs must be limited to specific characters


  • All communication between front & back end must occur over secure HTTPS connections
  • SSL/TLS certificates must be enabled on the server to verify authenticity of secure connections


  • Images must be stored securely using dedicated file storage solution
  • System to be put in place to guard and rotate secrets
  • Third party dependencies must be run through vulnerability scanners


What We Did Well


The research team's success building rapport in user interviews fostered a safe space for participants. Users were open and forthcoming in conversation and in turn, we were able to collect incredibly rich insights that informed foundational decisions across the platform.


SPI directors were impressed with the progress and quality of work we achieved. Clients felt that their feedback was understood and reflected in research and design decisions. They were especially excited about learning from research team's insights.


This project was an incredible opportunity to work in a highly collaborative team environment. We were applauded for working cohesively across teams and reporting our findings in concise and digestible deliverables in cross-functional team collaborations.

Client Feedback

"I am sure that this is going to be life-changing for so many people."
"It is just amazing seeing everything that we’ve been talking about, put into action. It’s so magical, I just cannot describe how happy I am."

What We Learned


Our participant pool was provided directly from the client and most of the TDA Graduates we spoke to held leadership positions as either SPI Country Directors or Wellness Team Representatives. Moving forward, it will be important for future phase research teams to speak to a broader range of users to build a more comprehensive understanding of needs within the community.


In Sprint 2, it became clear that there was a gap between client expectations and feasibility for the overall scope of the project. In an all-hands retrospective, the decision was made for each team to work with Product to discuss feasibility and scaling down scope to communicate our limitations with the client and refocus efforts on the most critical features. Future teams will benefit from having client check-ins early and often to manage expectations and keep project progress on track.

Get in touch

Please don't hesitate to reach out if you'd like to learn more about me or if you're interested in working together. Shoot me an email at