Getting dental claims paid faster

‍The Problem

Dental billing teams spend hours every day reviewing insurance claims before they're submitted. Copilot already helped offices organize that work, but every claim still required a person to decide whether it was ready to send.

The obvious solution seemed simple: automate claim submission. The problem was that our users didn't trust automation enough to let it make decisions on their behalf. During customer conversations, one concern came up repeatedly: "I don't want AI submitting claims I haven't reviewed."

The challenge wasn't building automation. It was building trust.

Understanding the Real Problem

Early conversations with billing specialists revealed something unexpected. Users weren't asking for a fully automated workflow. They wanted automation that behaved like an experienced teammate—handling predictable work while allowing them to step in whenever necessary.

That insight completely changed our direction.

Instead of designing a one-click automation feature, I began exploring how users could define their own rules for automation. The product shifted from an automation feature into a configurable workflow engine.

‍

My Role

I was the sole Product Designer responsible for the experience from discovery through delivery.

I partnered closely with Product, Engineering, and Customer Success to understand billing workflows, identify opportunities for automation, validate concepts with customers, and design the end-to-end experience.

Rather than designing isolated screens, I focused on creating a system that could scale across hundreds of providers, insurance carriers, and billing preferences.

‍

Key Decision #1

Control builds trust

Our earliest concepts emphasized automation. The more we tested the idea, the clearer it became that users didn't want software making invisible decisions. Instead of hiding automation behind a single toggle, I designed a layered settings model.

Users could enable Autopilot globally while still configuring behavior by provider, insurance carrier, procedure code, and timing. This gave offices confidence that automation reflected their billing policies instead of ours.

Key Decision #2

Make automation explain itself

Automation often feels unpredictable because users don't understand why something happened. To address this, I designed visibility into every automated action. Claims displayed clear statuses throughout the submission process, and users could easily identify which claims were being managed by Autopilot versus requiring manual review. Instead of wondering what the system was doing, users could always understand what happened and why.

Key Decision #3

Design for exceptions—not just happy paths

Insurance billing is messy. Claims are rejected. Carrier rules change. Offices have unique workflows. Rather than optimizing only for ideal scenarios, I designed flexible exceptions into the experience.

Users could exclude specific carriers, providers, or procedure codes, delay submissions by business days, schedule daily batch processing, and override automation whenever needed. The result was a system that adapted to different offices instead of forcing every office into the same workflow.

‍

Final Solution

Carrier Settings

Billing teams can customize automation independently for every insurance carrier, allowing offices to adopt Autopilot gradually instead of all at once.

‍

Time & Delay Settings

Submission timing is flexible, supporting immediate processing, business-day delays, or scheduled daily batches to match existing office workflows.

‍

Claims Queue

Claims clearly communicate whether they're awaiting review, actively managed by Autopilot, or require user attention, helping teams focus only on exceptions.