Conversational Healthcare AI

COMPANY

GoodRx

ROLE

R&D Design Lead

EXPERTISE

AI / Service Design

TEAM SIZE

4

PROBLEM

GoodRx's compartmentalized product and feature approach fosters confusion, undermining the product's utility beyond mere drug discounts. The search functionality, limited to prescription drugs, frequently falters due to common spelling errors and unfamiliarity with medication configurations like dosage and form. This mismatch often results in customers being unable to secure the correct coupon configuration, eroding trust when faced with unexpected additional costs at the pharmacy counter. Concurrently, pharmacies bear the brunt of this product inefficiency as they resolve customer discrepancies, adding to their frustration with GoodRx and savings coupons.

CONSTRAINTS

Focus on mobile web (most popular entry point)

Use the existing design system where possible

Small team

PRODUCT EVALUATION

To help executive leadership understand the current product's complexity, I created a service blueprint revealing the whole customer journey (it's unbelievable, but they never had this before). This journey involves ineffectively interacting with multiple people, various locations, and systems. Everyone involved faces challenges at many steps. I call this the coupon-first model:

SERVICE BLUEPRINT

A compartmentalized product and feature approach foster confusion, undermining the product's utility beyond mere drug discounts (below left). The search functionality is limited to prescription drugs (below middle).

Frequently, this falters due to common spelling errors and unfamiliarity with medication configurations like dosage and form (below left).. This mismatch often results in customers being unable to secure the right coupon (below right), eroding trust when faced with unexpected additional costs at the pharmacy.

Concurrently, pharmacies bear the brunt of this inefficiency as they navigate resolving these discrepancies with customers, adding to their frustration with GoodRx coupons.

SETTING A VISION

I came up with unconventional ideas to replace the coupon-first model with an automated system. I call this the Member-first model. This would remove the confusion of customer-configured coupons, checkout failures, and the pharmacist's having to troubleshoot GoodRx's product failures.

SERVICE BLUEPRINT

Along with the blueprint, I also presented the experience in storyboards to help capture the idea, the feelings, and the simplicity of the product interaction.

WORKING BACKWARDS

Working backward from the vision, continue to revolutionize access to GoodRx by breaking down its product silos and enhancing the user experience through intuitive and conversational AI. Expose GoodRx features available in context and help customers through their journey, meeting customers where they are and not by feature silos. (Let's ship a cohesive product, not the company org chart.)

Random customer data and AI identify the most common drug configurations within the customer's location. Contextually set the most common drug configuration on behalf of the customer. This helps successful claims at pharmacy checkout and lowers the likelihood of the pharmacist having to troubleshoot on GoodRx's behalf.

WIREFRAMES

If the customer visits GoodRx and knows what they are looking for (below), they can type it or say at and see holistic results that are contextual to the query (below middle), again breaking down the silos of GoodRx.

If the customer is unsure, they can start a conversational AI experience. This helps the customer get what they are looking for, builds confidence, and again exposes the full features of GoodRx in context.

GoodRx features are exposed to the customer upfront on the homepage. Tapping a decision (above left) starts the conversation with AI to help guide the customer (above right).

Using the keyboard or voice to tell the drug name. With drugs being so hard to spell, voice (above right) is great as you can just say it how it sounds, and the service will decipher what the real drug is.

Drug name is understood and the price is shown along with the pharmacy coupon and futher options below (above left). Choosing an option keeps the conversation going deeper into for example drug information (above right).

VISUAL MOCKUPS

I used the existing design system but also refreshed it with new components to create the conversational style.

OUTCOME

Working with our Insights team this work is currently undergoing A/B testing on the live mobile site with specific customer groups. The insights gained will directly inform future enhancements and optimizations. Then a broader rollout.