Valley Diabetes & Obesity automated 90% of eligibility and inbox workflows in 12 weeks, saving $149K per physician and recovering 8+ hours per staff member weekly.
Article text
A Diabetes Practice Achieved 90% Automation and $149K Annual Savings
Every morning, the front desk team at Valley Diabetes & Obesity started the same way. Log into Medicare. Check eligibility. Log out. Log into Medicaid. Check eligibility. Log out. Log into Blue Shield. Log into Anthem. Log into the IPA portal. Twelve different payer portals before the first patient walked through the door.
That ritual was consuming 12 to 15 hours of staff time every week. And it was the visible problem. The invisible problem was what happened when those checks got skipped, rushed, or delayed.
Denials. Write-offs. Revenue that vanished before anyone realized it was missing.
If you're running a specialty practice and wondering whether automation is worth the friction, we do free 30-minute discovery calls to help you figure that out.
The Hidden Cost of Manual Eligibility
VDO is a diabetes and obesity specialty practice in Central California. Not a massive health system. Not a venture-backed clinic chain. A real, independent practice dealing with the same administrative nightmare facing most small specialty groups.
The team was verifying coverage manually across 10 to 15 different payer portals every single day. Medicare. Medicaid. Commercial plans. IPA contracts. Each portal had its own login, its own interface, its own quirks.
The time cost was obvious. But the downstream financial impact was worse. When eligibility checks were delayed or incomplete, the practice didn't find out until the Explanation of Benefits arrived. By then, the patient had already been seen. The claim was already denied. The write-off was already inevitable.
According to the 2024 CAQH Index, the U.S. healthcare industry could save $42 billion annually by fully automating eligibility verification and claims-related transactions. VDO was living the manual version of that statistic.
The Deployment: One Workflow at a Time
The practice didn't try to automate everything at once. That's usually where small practices fail with AI projects. They scope a massive transformation, get overwhelmed, and abandon it six months in with nothing to show.
VDO took a different approach. They deployed Agentman's healthcare agent skills in phases, one workflow at a time.
Phase 1: Eligibility Verification (November 2025)
The first agent skill handled morning eligibility checks automatically. Instead of staff logging into 12 portals, the system ran those checks before the office opened. Coverage status, authorization requirements, and any red flags surfaced in a single queue.
Staff only touched the exceptions. The routine cases ran themselves.
Phase 2: Inbox Triage (January 2026)
The second phase tackled the morning inbox backlog. Clinical messages, administrative requests, and routine correspondence were automatically classified, routed, and prioritized.
The inbox that used to consume 2 hours every morning now took 20 minutes to review.
Phase 3: Prior Authorization (In Progress)
The current phase targets prior authorization workflows. For a diabetes specialty practice, this is the highest-friction area. Prior auth requests typically consume 14 hours of staff time per week. That's where the next wave of capacity recovery is coming from.
This is the same diagnostic-first approach we use at AutoSolve Labs . We don't start with technology. We start with the process map.
The Numbers at 12 Weeks
The results after 12 weeks of deployment:
Automation rate: 90% of eligibility and inbox tasks.
The agent skills now handle end-to-end processing for routine cases. Staff review exceptions, not volume.
Staff time recovered: 8+ hours per person per week.
The front desk and billing teams reclaimed nearly a full day each. That time went back into patient engagement, complex billing issues, and the work that actually requires human judgment.
Denial reduction: 65% fewer eligibility-related denials.
By catching coverage issues before the appointment instead of after the claim, the practice eliminated the majority of preventable write-offs.
Financial impact: $107,000 to $149,000 projected annual savings per physician.
The savings came from denial prevention, increased appointment capacity, reduced write-offs, and recovered staff time.
Inbox processing: 120 minutes down to 20 minutes.
The morning inbox ritual that used to consume the first two hours of the day now takes 20 minutes of review.
The Compound Effect
The phased deployment approach created compound gains. Automating eligibility didn't just save time. It created the capacity to automate inbox triage. And inbox automation created the bandwidth to tackle prior authorization.
That's the pattern that separates successful AI deployments from stalled projects. Each phase builds trust, proves ROI, and creates space for the next phase.
We've seen this same pattern in law firms automating client intake , staffing agencies accelerating candidate response , and property management companies streamlining lease renewals . The automation that works is the automation that compounds.
Why This Matters for Small Practices
VDO is not a large health system with a seven-figure IT budget and a dedicated innovation team. It's a small specialty practice. The deployment took weeks, not years.
The key conditions for success were specific:
Start with the highest-friction workflow.
For VDO, that was eligibility verification. For your practice, it might be prior authorization or referral tracking or appointment reminders. The workflow that causes the most daily pain is usually the one with the fastest ROI.
Measure before and after.
VDO tracked portal logins, staff hours, denial rates, and processing times from day one. That data made the decision to expand obvious instead of speculative.
Human-in-the-loop design.
The agents handle volume. Humans handle judgment. Staff didn't get replaced. They got promoted from routine checking to exception handling.
What You Should Check in Your Practice
If you're running a specialty practice, medical group, or any referral-dependent service business, ask these three questions:
How many payer portals does your team log into daily?
If the number is above 5, there's automation opportunity.
What's your eligibility-related denial rate?
If you don't know the number, that's the first problem.
How much staff time goes to routine verification vs. actual patient care?
Anything above 30% routine work means there's capacity trapped in manual processes.
You can run this diagnostic yourself. Pull your denial data for the last 90 days. Flag every denial related to eligibility, coverage, or prior authorization. Calculate the write-off value. Then calculate the staff hours spent on those workflows.
If you want help mapping the process and identifying the automation opportunity, book a free workflow call . We'll walk through it the same way we did for VDO, and we'll tell you honestly whether automation makes sense or whether the fix is simpler than that.