The research from this project was two-fold:
(1)Initial problem discovery research among two customer cohorts
(2)Diving deeper into usability to improve designs that performed poorly in testing
Initial Problem Discovery
Problem
Capsule’s old onboarding process was built for customers to transfer prescriptions at other pharmacies into Capsule. The onboarding process was a stepwise form that could only be filled out if a customer had prescriptions to be filled.
A survey uncovered that many customers who were looking to “get started” at Capsule did not have medications to send from another pharmacy, specifically these customers either didn’t have prescriptions or had prescriptions sent over via e-script from their doctor. The team looked to research this behavior among potential users to understand their needs and build a more personalized onboarding process.
Research Objectives
Understand why visitors without prescriptions want to “get started” with Capsule?
Understand why visitors who just sent in a prescription with Capsule want to “get started” with Capsule?
Methods
Individual customer interviews were conducted with 19 customers to understand the two types of customer cohorts:
Customers without prescriptions
Customers who already sent in prescriptions from their doctor.
Findings
The “no prescription” users and “doctor just sent my prescription to Capsule” users had some similar motivations.
Information finding. Users in both cohorts signed up with Capsule to find more specific information than was available on the logged-out website, (information on specific medications, more detailed information on Capsule’s prices, information on insurance and pricing).
Vetting Capsule. Users in both cohorts signed up with Capsule to vet Capsule as legitimate.
Capsule “How to” Confusion. A few users in both cohorts thought that signing up was required as a part of the Capsule process.
However, their behavior also differed in a few key ways:
“No prescriptions” users:
Are looking into Capsule prior to a doctor’s appointment or prescription event.
Are motivated to set up an account with Capsule to share important information ahead of time to avoid potential mishaps, delivery issues or delays.
“Doctor sent a prescription in already” users:
Are introduced to Capsule in their doctor’s office
Are motivated to set up an account with capsule because they didn’t understand how Capsule works and wanted to learn more.
Takeaways
Since motivations differed among the three cohorts, the team decided to create a trifurcated onboarding approach. Though the customers all saw the same account information once signed in, this approach allowed the team to bucket customers into these three groups for future marketing initiatives. The new onboarding designs were tested with poor results. Although the new onboarding was slightly better at driving sign-ups it drove down order conversion, prescriptions transferred in and showed no lift in account conversion. Further research was conducted to improve the design (see below).
Customers had questions around pricing /cost and largely wanted a way to vet Capsule (ie. Free delivery? What’s the catch?) that were not addressed in the more detailed logged-in account FAQs. Since these questions popped up across cohorts, we investigated a separate pricing page on the logged-out website (See Case Study #2, “How past research informed website updates to elicit trust among customers” for more) .
Design Improvements
Problem
The initial design update created performed poorly in A/B tests. The flawed designs were reviewed with actual users to test the usability of the new design and understand impacts on metrics and actions farther downstream from this process.
Methods
5 customers from the “no prescriptions” and “doctor sent my prescriptions” cohorts were interviewed about their experience with the original trifurcated design. Users were chosen among these groups who both did and did not send in prescriptions after creating accounts. Interviewing real users allowed us to understand the motivations behind their choices and make sense of the confusing downstream results.
Hypotheses
Customers were confused by the options in the onboarding flow.
The logged-in accounts did not have the right information to convince customers to convert.
We underestimated how long it would take customers with accounts to send in new medications.
Findings
The new set of onboarding options confused customers, especially the options to “sign up without a prescription” and “fill the prescription my doctor sent to Capsule”. The confusion came from:
Not understanding what it meant to be “with/without a prescription”
Not understanding what “situations” where a doctor would have already sent in a prescription
Customers wanted more information in the account than they were able to find in the FAQs. They wanted to understand insurance/pricing, information on their doctors, and more specific information about medications, inventory, and regulations.
There will be a subset of customers who take a long time to send because they have nothing to fill, they were just curious about creating an account. These customers can be targeted through marketing.
Takeaways
The product team used the feedback to update the design, most notably by updating copy and simplifying the options for customers. Since the three options were initially confusing the team focused on asking the customer to do one action or answer one question at a time to filter them into the correct group and the correct CTA.
Based on the feedback from customers, a few different variations were tested. The chosen variation (variation 3) for the new onboarding flow showed a notable increase in web-to-order conversion and additional accounts created.
