Shepherd
Universal Search
I helped reshape Shepherd.vet, a cloud-based veterinary practice management platform that allows practices to manage daily workflows, invoicing, and patient care from a SOAP-based dashboard (Subjective, Objective, Assessment, Plan). I led the redesign of patient forms, patient profiles, and an error 404 page. I also designed Universal Search to shorten the time needed to search for crucial patient information during emergencies.
IMPACT
I implemented a universal search that reduced time on task by 50% in the core flow
ROLE
Product Designer
TIMELINE
2 weeks
TEAM
1 Product Designer
1 Founder
1 PM
5 Developers
STATUS
SHIPPED
THE PROBLEM WE’RE SOLVING FOR
How might we efficiently provide vet techs with information they need, for a smooth emergency procedure?
THE CHALLENGE
Vet techs took 10 minutes sifting through screens to find crucial information piece-by-piece during time-sensitive emergencies.
The product held a lot of information, organized in a way users had to understand before they could find anything. Search was slow, and slow search meant users lost trust that the information was there.
User problem
Users needed specific patient information quickly, but retrieving it took around 10 minutes sifting through each page to get to the desired content with important patient information.
Product challenge
The search had to work across the software without making users re-learn the underlying structure, or disrupt the product’s variety of interactions.
The Impact
Optimized Core Workflows
Streamlined the primary journey to eliminate redundant inputs and cut time-to-value exactly in half.
Shifting User Sentiment
Transformed customer anxiety into high confidence by introducing transparent system feedback and proactive validation.
Secured Subscriber Retention
Protected recurring revenue by re-engineering post-purchase touchpoints to continuously highlight ongoing product utility.
RESEARCH & DISCOVERY
How did the users feel about the process of finding important information?
Looking into the information architecture and feedback, tying them together with our business goals was our priority.
Information architecture
Reviewed how information was organized across 20+ practices and where that structure forced users to know the path before searching.
User feedback
Collected feedback on what made finding information slow or unreliable.
Goals
Defined what search needed to achieve for users and for the team.
Accessibility
Considered accessibility in how search was entered, read, and navigated.
DEFINE
Why did we land on a universal search?
Previously, there was a search that my team noted did not provide value due to the type of information it presented. So, after hearing what the vet techs reported as needed, the need to cut down our information architecture for how much time our core flows took, and considering our sprint timeline we decided a universal search would provide the most upfront convenience.
ITERATIONS
Exploring ideas of how our search engine could fit in.
How can a user find the right information without first understanding how it is organized?
SOLUTION
Final Solution
I designed several iterations and proposed them to engineering, the PM, and the founder. We shipped the founder's version, a mix of interaction approaches from the options.
MY REFLECTION
What I'd do differently
with more time and clearer direction.
This was fast, ambiguous, and cross-continental. Here's what I learned — and what I'd push harder on next time.
Doing user tests. [explanation]
Go through more explorations. [explanation]
Have more discussions with the developers on how we could piece-by-piece ship something more robust. [explanation]
Do a deeper dive into the Information Architecture. [explanation]