In 2022, my team was focused on helping Solv become a go-to destination for convenient care, and one of our goals was improving our users' search for care.
‍
The Search Results Page (SRP) plays a big role in the booking experience on Solv, a healthcare marketplace primarily for same-day and next-day visits. It’s a way for us to surface providers to people who need care: As many as 20% of people who book on Solv touch the SRP.

But the design was outdated, and the experience wasn't optimized for an efficient, helpful search (frustrating for anyone, but especially for those in need of healthcare).

I led a refresh of the SRP, updating the page's overall look and feel, with the primary goal of increasing booking conversion.

The SRP is a moment of truth for our user, especially ones who have never used our product before. Their experience on the SRP can determine whether they choose Solv for their healthcare booking needs now and in the future.

As lead designer, my objective was to uncover what exactly is wrong with the SRP in its current state, so that we understand what problems we're trying to solve and where we might have the greatest impact on booking conversion from the SRP.

Since the last update to the SRP was back in 2020 to include COVID-related UI, it was clear that certain improvements were just must-haves by now: We know it's the right thing for a search experience to include sort and filter options; we know that outdated, irrelevant info on the SRP was causing a lot of clutter and was distracting and potentially confusing to our users. Now was our chance to address these issues—but also to uncover any new opportunities to improve.

Within a sprint, I conducted a heuristic evaluation, user interviews, a data deep dive, and a direct/indirect competitor analysis of the search-and-book experience.  

From here, I clustered the research findings into themes, and my PM and I aligned on two major pain points we hoped to address with this project.

When searching for care on the SRP, our user...

To kick off my exploration on solutions, I led a design jam with my design director and PM to go through the current state of the page and its UX. I wanted us to get on the same page about our assumptions and thoroughly ensure our hypotheses were not just addressing the two key pain points, but impactful towards our goal of increasing bookings.

I asked:

This exercise was helpful to identify the visual clutter and table-stakes improvements that we as a team wanted to work on. It was also a useful tool in defining the requirements for this project—what exactly is in and out of scope—in collaboration with my PM and tech lead.

Ultimately, we converged on four main hypotheses around how we might improve our user's search experience so that she is more likely to book with a provider.

I bucketed these hypotheses into two categories in order to help me understand the work that needed to happen: UI-focused and UX-focused. Within each bucket, my PM and I listed our must-have and nice-to-have improvements, which were prioritized and de-scoped along the way.

One other important piece of this project not mentioned above was bringing the SRP's look and feel up to speed with Solv's design system, presenting our users with a cohesive, seamless experience across the Solv product.
‍
Diving into the solution space, I looked back at competitive research to see what patterns are most familiar in the search-and-book experience, sketched out ideas, and then jumped into Figma to see what's working and not working with our design system components.  

Over the course of several checkpoints (concept reviews, design crits, an eng review, and finally a product review), I presented designs, recommendations, and iterations to various stakeholders. Eventually we landed on a direction for the SRP that we felt confident about.

Below I'm highlighting three of the updates we made in the designs.

One common pattern we see on SRPs in marketplaces are badges or icons that can quickly convey pertinent info about a search result. These small-but-mighty indicators can be helpful for a consumer in making a decision.

We looked at historical research about what information our users say is important to them when they search for care and the main reasons people choose a provider.

Outside of ensuring that a user on our SRP could easily find this information about their search results, we wanted to incorporate icons on the location cards when a clinic exemplified these specific qualities.  

Our icon library didn't have workable options for these badges at the time, so after determining which qualities are most important for this feature, I designed the iconography for our badges. We called them accolades.

(Click on this image below to enlarge and see which accolades made the cut).

In terms of improvements to the current UI, the way we were showing virtual visit (telemed) options felt more like an ad, less like an actual option for care that would compel a user to click and book.

With the refresh, we wanted to show users all their options for care in a standard, more intuitive way. The goal of this new design below was to improve the user's expectation of what's behind the "Video visits" toggle (more time slots, i.e. more ways to get care now).

(Click on the image below to enlarge).

The way filters worked in the existing implementation was very confusing to users because what we called "filters" on the SRP weren't actually true filters—they were search parameters. We saw little to no engagement on this button, or the chips below.

With this refresh, we cleaned up this UI and updated the filter button to display actual filters that users can apply to refine their search results. Though basic for now, the goal here was to provide our users with a working filter feature, and then be able to scale our filters as we grow our offerings and continue learning from usage.
‍
(Click on the image below to enlarge; note that since this refresh, we've separated the sort and filter function, among other other UX/UI updates).

At this point, I'm feeling pretty good about where we've landed with the designs, after implementing feedback and getting buy-in from teammates and stakeholders at each checkpoint.

As a startup, we ship fast and learn fast, so testing designs prior to launch doesn't always happen (or need to, truthfully). But with such a prominent surface and step in the booking flow on the line, testing the designs could give us a little more confidence about our direction.

Fortunately, we were able to immediately test out our new filter offerings and accolades on Solv's Directory pages, the SEO pages where top-of-funnel users land after searching, say, "urgent care near me" on Google. We wanted to see how users were interacting with these new features, and we received early positive signal (a strong correlation between users who were applying filters and then going to make a booking) by monitoring these additions.

I also conducted a series of rapid usability tests on the new designs using Maze.

For this, I created prototypes of the designs and tested them against the current design with 25 testers each. I wanted to look for overall improvement in usability between the current and new design, specifically testing the things we want our users to be able to accomplish on the SRP. 

We were happy to see improvement on each of the tasks (which is great, however it’s always good to keep in mind that we can see something different when it launches and it’s out in the wild). But at this point, it served as a nice sanity check for us.

Here's a snapshot of the results. These usability scores are measured by different factors including the time it takes to complete a task and amount of misclicks on the screens. 

(Click to enlarge the images below).

We rolled out our refresh in phases using A/B testing. Ultimately, our final hypothesis was that by updating the SRP UI, surfacing video slots on the SRP, and adding the ability to filter results, our user will be more likely to book with Solv and have an easier, more intuitive, and more delightful experience.

After our testing, we found that the refresh outperformed the current SRP, and in November 2022, we rolled out the new experience to 100% of users—just in time for flu season.

We continue to learn, iterate, and improve on the SRP experience. You can see the live experience here.