Autonomous collections at *marketplace scale.*

Clipboard Health operates a healthcare staffing marketplace that connects healthcare facilities with nurses and other healthcare professionals, filling shifts on demand across thousands of facilities nationwide.

The marketplace model has powered rapid growth, but it also created a billing operation of enormous scale: every filled shift generates an invoice, and every facility is an account that must be managed, followed up with, and collected from.

By 2026, Clipboard Health was carrying roughly $110 million in accounts receivable spread across more than 5,000 facility accounts. At that volume, collections is not a task, it is a full-scale operation. Every promise to pay had to be logged by hand, every dispute chased down across email threads, and every customer inquiry routed manually to the right owner. The cost to service the AR book kept climbing with the business: more accounts meant more inboxes to monitor, more follow-ups to schedule, and more collectors spending their days on repetitive administrative work instead of resolving the accounts that actually needed human judgment.

The process worked, but it did not scale, and the manual cost to service the receivables book was eating into the economics of the business.

Clipboard Health needed a solution built for volume, one that could operate across thousands of accounts simultaneously without adding headcount, and one that could understand what customers were actually saying rather than just sending reminders on a timer.

Daylit's platform deploys autonomous AI agents that listen inside customer inboxes. When a customer replies, the agents automatically detect what kind of issue the message contains, a promise to pay, a dispute, or an inquiry, and automatically create a case for it.

For a team managing 5,000+ accounts, this changed the fundamental unit of work. Instead of collectors reading every reply and manually triaging what to do next, the platform did the listening, the classification, and the case creation on its own, leaving the team to handle only the exceptions that genuinely required a human.

• Detect every reply as a promise to pay, a dispute, or an inquiry, the moment it arrives
• Case & sequence each issue automatically, following Clipboard Health's own SOPs
• Act autonomously, following up on payment dates, escalating disputes, answering routine inquiries

Work that once consumed the collections team's entire day, reading replies, logging promises to pay, opening dispute cases, scheduling follow-ups, now happened automatically, in the background, across the whole book of accounts at once.

Because every customer reply was detected and acted on the moment it arrived, nothing sat in an inbox waiting to be read. Promises to pay were tracked and followed up on their promised dates. Disputes were cased and routed immediately instead of aging in a thread.

For a business carrying $110 million in receivables, faster collections translated directly into working capital.

But the bigger change was structural: Clipboard Health's collections capacity is no longer tied to headcount. The team that once spent its days on manual triage now focuses on the high-value accounts and complex disputes where human judgment matters, while Daylit's agents run the volume.

As the marketplace continues to grow and the account base expands, the collections operation scales with it, without the manual cost that used to grow alongside it.