Key Takeaways:
Every missed call and every declined service that never gets a follow-up adds up to real money walking out the door.
Most managers only hear a tiny fraction of their own BDC calls, which means most of what's happening on the phones stays completely invisible.
The AI tools worth adopting know exactly when to hand a conversation to a person, and they get graded on the same rubric your own team already uses.
Corbin Witte, GTM at Avoca, opened his AI Show and Tell session with a claim that stands out at an AI conference: fixed ops runs on relationships and loyalty more than any other part of a dealership. And it holds more untapped revenue than any other department, too.
At this year’s AutoIndustry.AI Summit, Witte walked through Avoca's platform live and letting the demo carry the argument.
Avoca built its foundation in home services — HVAC, plumbing, and roofing — where the platform now runs 75% of the top 25 private-equity-backed operators in North America and recently crossed a billion-dollar valuation. The company now brings that same system into dealership fixed ops.
The Real Problem: Money Slipping Through Everyday Gaps
Witte framed the core issue as a technology gap hiding inside routine daily operations:
Calls during the morning rush go unanswered. A busy service drive means some calls simply never get picked up.
Declined services from last month's multipoint inspections sit untouched. A customer says "not today" on a brake job, and the follow-up that would bring them back rarely happens.
BDC conversations stall right before booking. A customer says they'll think about it, and no one circles back to close the loop.
None of these gaps stem from a fixed ops problem. Each one traces back to a technology gap, which makes it solvable with the right system in place.
Handling the Routine Calls Cleanly
Avoca's core call-handling demo showed a customer calling in during a busy morning. The AI pulled the customer's record from the DMS in real time, surfacing their vehicle, service history, and any open recalls, then asked what brought them in, cross-referenced actual calendar availability, and offered real appointment slots. Routine work — oil changes, tire rotations, recall completions — gets handled cleanly and completely by the AI.
Knowing When to Bring In a Person
Witte drew a clear line around what AI handles well and where it needs backup. An AI system can acknowledge a customer's frustration, apologize, and respond with real empathy, and it still runs into limits with a nuanced billing dispute or a misdiagnosis conversation. Avoca built a human-in-the-loop team directly into the platform for exactly that reason: 80 live, trained agents available for transfer into any call, 24/7, currently used on roughly 8–12% of all calls running through the system.
Witte offered a direct test for evaluating any AI voice vendor making bigger claims: ask to see the containment data. A vendor claiming their AI handles every call without human backup deserves real scrutiny on that number specifically.
Coach: Grading Every Call Against Your Own Standards
Most managers hear a small slice of their own BDC calls each month, typically somewhere between 3–5%. That leaves the vast majority of daily phone activity completely unreviewed, whether that's reassuring or concerning depends entirely on how confident a manager already feels about their team.
Avoca's Coach feature listens to every call — BDC, service advisors, and sales — and scores each one against a rubric built specifically with that dealership, reflecting what that store actually values in a conversation. In one example Witte walked through, a customer called to schedule a brake job, and the BDC agent stayed polite and answered questions, and the customer still hung up saying they'd think about it. Coach flagged the call at 54 out of 100, identifying three specific missteps: the agent never confirmed urgency, never asked when the brakes were last serviced, and never offered a specific time slot. Within 15 minutes, the fixed ops director received an alert recommending a callback. The store called back, and the customer booked for later that week.
Coach surfaces patterns for a manager to act on. The decision to retrain, restructure, or make a personnel change stays entirely with a human.
Nurture: Turning Declined Services Into Booked Appointments
Avoca's outbound Nurture feature targets a specific and common leak: every multipoint inspection surfaces declined services — brakes, tires, fluids, filters — and most of that revenue evaporates simply because no one follows up. Closing that gap typically takes six or seven touch points.
The system pulls declined line items directly from the DMS and builds a nurture campaign around them. A customer who declined rear brake service gets a message two weeks later flagging the specific service and offering an open slot. When the customer confirms, Avoca books the appointment directly into the DMS and sends confirmation, with no agent involved at any step.
Nurture ranks among Avoca's highest-ROI products, since the revenue tied to declined services would otherwise drop to zero. One home services client recovered $800,000 in a single season using this exact approach, and Witte described the fixed ops opportunity as comparable in scale.
Timing makes or breaks a nurture campaign:
60–90 days out counts as too late — customers move on and forget the conversation ever happened
Two to six weeks post-visit hits the sweet spot, giving customers enough time to reconsider without losing the thread entirely
Three Questions That Cut Through Any AI Pitch
Witte closed with a framework built for the rest of the conference floor, useful for evaluating any AI vendor, Avoca included:
Does it book natively into your DMS, or is there a manual step somewhere? A human retyping AI output into a system like CDK counts as expensive data entry, not automation.
What's the containment rate on hard calls? Upset customers and complex questions reveal how a system actually performs under real pressure, not staged demo conditions.
What does onboarding actually involve, and how long until real data starts showing up? Fixed ops AI requires configuration, call flow logic, rubric calibration, and DMS integration testing. A vendor promising a fully tuned system within 48 hours is setting an expectation well ahead of what the technology can currently deliver.
What Came Up in the Q&A
A few audience questions rounded out the session with useful detail:
Customer comfort with AI has shifted fast. Witte noted that hesitation toward speaking with AI was real a few years ago, and voice quality has advanced enough that most customers care only about getting the issue resolved.
Dealerships can route calls to their own BDC team, keeping Avoca's human-in-the-loop staff available separately. Many dealerships use Avoca's trained agents specifically for after-hours coverage, while keeping their own team in the loop during business hours.
The rubric grading system works both at the individual and team level. A manager can review a single rep's scores over time — booking rates, empathy, follow-up on objections — or see the full team's performance side by side.
The Bigger Picture
Witte's framing throughout the session reduces to one point: the revenue lost to missed calls, unreviewed conversations, and forgotten declined services already sits inside a dealership's own data. The technology exists to surface it, book it, and follow up on it automatically, and the dealerships getting real value from it are the ones asking sharp questions about containment rates, DMS integration, and onboarding timelines before signing on with any vendor making big claims.
