Increasing Patient Conversion

Client Overview

Talkiatry is a national mental health organization offering accessible psychiatry and therapy services through virtual appointments. Working in-network with major insurers, Talkiatry connects patients to licensed psychiatrists for evaluations, personalized treatment plans, and medication management — combining technology with a patient-centered philosophy to make high-quality mental healthcare widely available.

The Problem Wasn't Insurance. It Was Trust

Every patient who calls Talkiatry is already at a vulnerable moment — reaching out for mental health care, often for the first time. What happened next in that call shouldn't have mattered as much as it did.

Behind the scenes, Talkiatry's insurance verification process depended on a manual team working through complex, behavioral-health-specific coverage rules one patient at a time. The accuracy of that process — just 70% — meant patients were often given the wrong picture of what they'd owe, or left waiting while staff tracked down answers. Appointments got delayed. Claims got denied after the fact. And patients who'd worked up the courage to seek care were sometimes lost in the gap between "I want help" and "I know what this will cost me."

For Talkiatry, this wasn't a back-office efficiency problem. It was a patient experience problem with real clinical stakes, and a growing operational cost with no ceiling as the organization scaled.

Sohar's Approach: Solve for the Patient First, and the Business Follows

Talkiatry didn't come to Sohar asking for a piece of software. They came with a harder problem: how do you make insurance complexity disappear for the patient, without asking your team to absorb more manual work to make that happen?

Sohar's approach started with what makes behavioral health different. Carve-outs, Managed Care Organization structures, and payer-specific nuances in X12 service codes mean that a generic eligibility request often gets behavioral health wrong in ways that don't surface until a claim is denied weeks later. Getting it right upfront — not catching it downstream — was the design constraint from day one.

That meant building toward three outcomes simultaneously, not sequentially:

• For the patient: faster, more accurate answers about coverage and cost, delivered before the anxiety of the unknown could become a reason to disengage.
• For the operations team: a shift away from manually chasing payer information toward a system that got more accurate over time.
• For the business: fewer denials, lower cost-to-serve, and a funnel that converted more of the patients who'd already decided to seek care.

Sohar also integrated with Talkiatry's billing vendor, Candid Health, to ingest adjudicated claims data — allowing the system to dynamically improve its own accuracy as real payer behavior came in, rather than staying static after go-live. That meant Talkiatry's engineering team never had to own or maintain the integration themselves.

Implementation followed the same patient-first logic: rather than rushing to full production, Sohar and Talkiatry agreed the system would go live only once benefits data accuracy exceeded 90% across all payers, and 95% for the highest-volume ones — protecting the patient experience the whole effort was designed around, even if it meant a longer runway to full rollout.

The Outcome

The result wasn't just an efficiency gain — it was a funnel that finally matched the seriousness of the moment patients were in. Conversion improved by 18%. The operational cost of manual verification dropped 90%. And Talkiatry saw an 8:1 return within the first year, not because Sohar optimized a workflow, but because getting the first interaction right changed how many patients made it all the way to care.

How It Works, If You're Curious

Talkiatry's results were powered by two products working together from Sohar's platform — Discovery, which identifies a patient's coverage without requiring them to provide insurance details upfront, and Verification, which confirms eligibility and the specifics of their benefits. But the products were in service of the strategy above, not the other way around: the goal was never "implement eligibility automation." It was "make sure no patient falls through the cracks between wanting care and getting it."