Answered Calls Aren’t Booked Appointments: 10 Questions I’d Ask Before Buying Healthcare Voice AI
I’ve sat through many voice AI demos, and they all look great. The agent answers instantly, sounds natural and handles the call. Then I ask, “What happened to the schedule?” That’s usually where the conversation gets quiet.
Most buying guides tell you to score vendors on containment, the share of calls resolved without a human. That’s a call center metric, and nobody in your boardroom is paid on it. Your organization sells provider time, and an empty 5 p.m. slot is revenue that expires, like an airline seat after takeoff. An agent can answer 100% of your calls while your schedule sits at 81% full.
So don’t ask whether the call was answered. Ask whether the right patient landed with the right provider, and whether the slot stayed filled. That’s capacity optimization. Here’s the math: ten providers, 20 slots a day, 240 clinic days gives you 48,000 slots. Moving from 81% to 98% utilization recovers 8,160 visits, roughly $1.63 million a year at an illustrative $200 per visit. Our customers run 98% full schedules and a 2.2% no-show rate against the 18 to 20% national average.
Four questions about capacity
- What will you do to our utilization, and will it be in the contract? Ask for fill rate, revenue per provider day, third next available and no-show data from a named customer in your specialty.
- Can the agent see capacity across every provider, location and visit type? Most just offer the first open slot. Ask the vendor to show live how it chooses between three valid ones, and whose rules decide.
- What happens in the first hour after a slot opens? Cancellations are your most perishable capacity. Ask for the share of cancelled slots rebooked last month.
- Can it create demand where you have room? Inbound AI waits for the phone to ring. Ask for outbound work aimed at open slots, with attribution from ad click to completed visit.
Six questions about everything around the schedule
- What about demand that never calls? Web, chat, text and fax matter too, and an estimated 40% of referrals are abandoned before a visit. A voice-only agent optimizes one door of a building with five.
- Is every booking financially clean? A filled slot that ends in a denial isn’t optimized capacity. Look for eligibility checks at booking and copays collected before the visit.
- Who owns the rules after go-live? If every change is a vendor ticket, your logic will always trail your schedule. Your own team should edit rules and guardrails.
- Does it read and write to your EHR in real time? Cached availability causes double bookings. Ask for a reference customer on your exact EHR.
- Where does your PHI go? Require a signed BAA before any pilot, a sub-processor list and a ban on training models with your data. Ask how the vendor tests for hallucination and when the agent escalates.
- What does it cost when it works? Per-minute and overage pricing charges you more as you grow. Ask for the total cost at twice your volume, and for an exit clause if utilization targets are missed.
Change the scorecard
Replace containment with schedule fill rate, appointments booked with completed visits and revenue per provider day, and minutes used with a fixed total cost. That’s what your CFO will ask about a year from now.
How we answer at Steer Health
We built our growth product, FastTrackCare, around capacity rather than calls. It combines omnichannel capture, referral automation, waitlists, reactivation, and eligibility and payments, with the Luna voice agent as one channel among them. Over 92% of copays and unmet deductibles are collected before the visit. Rules live in Steer Studio, so your staff and IT control them. We offer 36 EHR integrations, including Epic, Oracle Health and athenahealth, with fixed costs and no overage fees. We’re also a member of the Coalition for Health AI. Across 109 organizations, customers average $3.5 million in incremental net revenue. We’re proud to have won the 2026 Hearst Health Prize.
Ask us these ten questions too. Vendors should be able to answer them.
Read the full guide: 10 Questions to Ask Before Buying Healthcare Voice AI
Comments
Post a Comment