Connecting people with care

Context

NOCD provides specialized treatment, which means we can only treat people who have OCD symptoms. Sending someone into treatment who isn’t a fit wastes their time, their therapist’s time, and company resources, so identifying the right candidates early matters.

That’s why the first step is a call with an intake specialist. They work through insurance coverage, screen for OCD symptoms, and match the person with a therapist, which also depends on location, since the match only works if an eligible therapist is nearby. The scheduling questions before the call do part of this screening.

The design problem: how do you guide someone through a complex onboarding without losing them along the way?

What changed

Directory and therapist profiles

The therapist profile is where people see credentials, and it’s where trust begins. It’s also where people see the clinical standards behind the care. I worked with marketing and our growth lead, who pulled data on which pages earned the most clicks and converted best. Profiles came out on top, since that’s where the human connection is. So we designed the directory to keep people from browsing endlessly and move them toward opening a profile sooner.

The profiles belong to one cohesive design system. I built them from the member profiles, which had recently been redesigned, reusing their UX patterns and components, with a UI designer supporting the visual work.

Asking just enough

Before the first call, we need to know where someone is and how likely they are to be in network, without being able to confirm it yet. Every question we add is another reason to leave, so the flow asks only what the intake specialist needs to make that first call work. I designed the UX for the full book-a-call flow and coordinated delivery with a UI designer, who handled part of the UI before I took on more of it myself.

We also tested letting everyone book a time on their own, without the call. It did worse: more people booked who weren’t eligible, or didn’t book at all. So self-booking opens up only where testing shows it works.

A second chance

The longer the gap between someone asking for help and speaking with a person, the less likely they are to complete a session. So when someone leaves the flow, we offer a second path, and which one depends on where they leave. The fallbacks came out of my work with the growth lead, and went through A/B testing.

Outcome

Both lifts came from the profile redesign and are statistically significant, measured before and after launch. The fallbacks bring in a few more conversions a day from people who would otherwise have dropped off.