Culina Health
Project Overview
Culina Health offers personalized virtual healthcare, connecting patients with top-tier Registered Dietitians (RDs) who bring diverse cultural expertise to their practice. Covered by most major insurance providers, the platform makes expert, accessible care available from the comfort of your own home.
As the Lead Product Designer at Culina Health, I led a comprehensive rebranding initiative to better reflect the company’s inclusivity and the high quality of care it provides. This involved collaborating with an illustrator to develop new brand assets, redesigning the company website, refining patient intake flows, and elevating visual and marketing materials across touchpoints.high-quality
A key feature I researched and designed was the RD filtering and matching tool. To align with Culina Health’s commitment to delivering truly personalized care, we revisited the patient–RD matching process to identify opportunities for improvement.
User feedback collected through email surveys revealed a need for more nuanced filtering and matching capabilities, with the goal of increasing patient satisfaction and reducing time to care. Using Maze for research and testing interactive Figma prototypes, we refined the filter experience based on insights gathered from users. The final launch of the feature enhanced the overall patient experience and strengthened the connection between patients and their ideal dietitian.
RD Filtering and Matching Feature
Problem: Over a 7 day period; we lost 36 patients on the matching screen. In addition, 40% of RD complaints regarding the intake process mentioned patient/RD mis-matching, suggesting that patients choose the wrong reason for visit.
Hypothesis/Solution: One potential solution to improve matching success is to change the matching format. The new format would provide patients with the option to choose more than one reason for visit (as it can be difficult for the patient to know which condition should be their primary focus), and filter RDs on additional criteria. Patients would also be able to read and sort through RD bios and choose the RD who they feel would be the best fit. This filter could also be added to our website as an alternative booking option.
Expected Outcome: Improved retention & custom satisfaction scores due to better matching, as well as a reduction in customer service tickets.
Research
We engaged with the Clinical Care team to gain valuable insights into the primary factors influencing individuals seeking re-matching with a different Registered Dietitian (RD) or expressing dissatisfaction with their match through customer service outreach.
After careful analysis, we identified key matching criteria, including:
✔️ Primary reason for the visit
✔️ Secondary reason(s)for the visit
✔️ Counseling style.
Additionally, we recognized certain 'nice to have' criteria such as diet preferences, languages spoken, and session availability.In exploring potential filter designs, I extended my research beyond healthcare companies, examining various sectors including e-commerce.
This broadened perspective aimed to integrate effective and user-friendly elements into our platform. An important l consideration in this process was to create a filter design that minimizes the need for significant adaptations between the desktop and mobile versions. This strategic approach was vital due to limited engineering resources, ensuring seamless functionality across different devices.
After testing several design approaches, I ultimately chose a collapsible drawer filter menu with expandable categories, which are essential for managing extensive options like “Reason for Visit.”
The design aimed to help users find criteria quickly without excessive scrolling while showing how many Registered Dietitians (RDs) matched their selections in real time. This transparency allowed users to understand the impact of each filter and adjust their choices with ease.
Common across familiar platforms, this filtering pattern offered both usability and scalability, ensuring a consistent, intuitive experience on desktop and mobile.
User Testing
Using the Maze platform as our research tool, we decided as a team we wanted to focus on gathering answers to the following questions:
✔️ What other criteria do you wish was in the filter menu that you could use to filter results?
✔️ Imagine you were going to see RD, what criteria would you use to find someone you felt was a good match?
✔️ After people are initially matched with an RD, are they using the filter?
✔️ Do you feel like the filter is helping you get to a better match for your condition and lifestyle?
✔️ Which filter criteria do users feel is most helpful?
✔️ Is the navigation clear to the user?
We created a Figma prototype in order for the users to complete the flow and asked a series of multiple choice and open-ended questions.
Findings & Insights
RD Matching Insights:
Survey participants rated comfort with match adequacy based on RFV and counseling style at an average of 7.6 Survey participants rated comfort with match adequacy based on additional filters at an average of 7.4
⭐ Takeaway: Participants felt equally comfortable with both methods of generating a match.
Survey participants rated comfort with match adequacy based on RFV and counseling style at an average of 7.6 Survey participants rated comfort with match adequacy based on additional filters at an average of 7.4
⭐ Takeaway: Participants felt equally comfortable with both methods of generating a match.
Booking Habits Insights:
Majority (68%) of participants indicated that they would choose RD who is exact match, rather than RD who fits RFV and counseling style alone. Minority (8%) of participants would book with RD with earliest availability, despite this being the highlighted option. 29% of participants were not immediately sure of who they would choose, and would have to further evaluate bios, etc to make a decision.
⭐ Takeaway: Participants are actually more likely to choose better match compared with earliest availability, ~1/3 of participants would utilize extra features to refine match and choose RD accordingly.
Navigation Insights:
Majority (69%) of participants are immediately drawn to the highlighted RD on the top of the page. 15% of participants notice matching percentage upon first look on page, and 15% of participants first noticed something else on the screen.
⭐ Takeaway: Participants will notice the highlighted RD first; highlighted RD should be strategic option. Since the main priority of adding a filtered screen is to improve patient/RD match, the best match RD should be presented first. Getting patients into care more quickly should be tested for separately as part of entire workflow testing.
Other Overall Insights:
While some/many participants will move through the flow quickly, based on top presented RD, approximately 30% or more will spend time exploring RD bios and adding additional filters to improve their likelihood of a better match.
The approach screen remains confusing (many participants did not complete this screen), and PM feels it is difficult to understand whether or not RD meets criteria for style preference based on primary page view.
The approach screen will be removed from flow, and approach will be added as a filter option Participants are looking for more “proof of care” (i.e. patient reviews, RD ratings, credentials, etc).
Participants are looking for more “humanization” of RDs (i.e. videos, more info on hobbies, etc).
Design Implications
The findings of our study will be integrated into the final design in the following ways:
✔️ The best match will be highlighted on top as top match.Additional matches will be sorted by soonest available.
✔️ Each RD will have 3 static specialities (will be presented in parent category).
✔️ Add credentials and “hobbies” for RDs.
✔️ Broader filter categories with sub-categories.
✔️ Sub-categories displayed on main RD listing (max of 3, always the same).
✔️ Profile will display specific conditions.
✔️ Profile will show filter criteria not matched by that RD.
RD Filtering & Matching Screens: Before/After
Opportunities For Further Iteration
✔️ Proof of Care and Humanization: Responding to participants' desire for more "proof of care" and "humanization" of RDs, explore the integration of patient reviews, RD ratings, and credentials. Additionally, consider adding elements like videos and additional information about RDs' hobbies to provide a more comprehensive and personalized view.
✔️ RD Profile Videos: Allow RDs to customize their profiles with a short introduction video. This customization can contribute to a more personalized and tailored matching experience.
✔️ User Education: Since some participants may not complete certain screens or find them confusing, implement in-app guides or tooltips to educate users about the features and encourage exploration. This can enhance user engagement and ensure that users make informed decisions during the matching process.
✔️ Continuous Testing and Iteration: Based on the insights gathered, continue to conduct user testing and iterate on the design. Test different variations of the filtering process, visual elements, and feature placements to identify the most effective and user-friendly approach.
✔️ Implementation of Additional Filters: Consider adding filters based on RD credentials, patient reviews, and session availability to provide users with a comprehensive set of criteria for making informed decisions. These additional filters can contribute to better matches and increased user satisfaction.
✔️ Collaboration with Clinical Care Team: Maintain an ongoing collaboration with the Clinical Care team to gather further insights and ensure that the filtering criteria align with the clinical considerations and preferences of both patients and RDs.
✔️ Measuring Impact: Implement tracking mechanisms to measure the impact of the new features on user behavior, retention rates, and customer satisfaction. Analyze the data regularly and use it to inform further refinements and optimizations.
By exploring these next steps and ideas, we can further enhance the RD Filtering & Matching Feature, providing patients with an intuitive, personalized, and efficient experience in finding the best-matched Registered Dietitian for their needs.
Research, Scoping, & Design for Native Mobile App
At Culina Health, I led the research, scoping, and design of the company’s first native mobile app.
You can review a walkthrough of the process here.
