• This thread is just the tip of the iceberg.The people ahead of the curve aren't Googling for answers — they're already in here, having the conversations you haven't found yet. DealerRefresh is free.Get the full picture →

500,000+ Dealership Calls Later: Here’s What We’re Learning About AI

Nov 11, 2026
1
1
Awards
1
First Name
John
At Matador AI, the most impactful upgrades come from fixing everyday dealership friction.


Drawing on my experience deploying enterprise AI for Fortune 500 companies, I work closely with Loris, our Senior Performance Manager, who brings over a decade of telephony expertise. Together, we’re building AI that reflects how dealerships actually operate and helps turn more calls into appointments.


Here is what we’ve rolled out recently and the data behind it.


1. Making AI Conversations More Natural​


When reviewing the phone recordings with BDC teams, we consistently heard how frustrating awkward silences, clunky robotic interruptions, and repetitive caller questions can be. We upgraded our voice engine to handle natural pauses, multi-part questions, and mid-sentence interruptions. Background inventory searches now run silently while the AI continues speaking.


Our recent before-and-after call analysis showed:


  • +12% caller name capture rate.
  • +6% recognition rate for returning service customers.
  • 21% fewer sales callers asking for a human before engaging with the AI.

2. Keeping Sales Opportunities Alive​


In our conversations with sales managers, their biggest concern was lost momentum when an advertised vehicle was already sold. Call AI now pivots automatically: it scans live inventory, asks for preferences, and suggests matching alternatives.


If a department doesn’t pick up a transfer, the AI steps back in to assist rather than dropping the call. Crucially, we also saw a 15% increase in post-call SMS follow-ups, keeping leads warm for the sales floor after they hang up.


3. Making Service Scheduling Easier​


Working alongside service directors, we watched their advisors get constantly bogged down by routine schedule modifications and tangled shared family contacts. Call AI now checks live lane availability, writes directly into supported schedulers, and accurately disambiguates shared contacts.


The result: a 14% increase in successfully completed service rescheduling requests without advisor intervention. Since June, Call AI has written nearly 15,000 appointments directly into dealer schedulers.


4. Built Around Actual Store Operations​


Dealers made it clear from day one that any automated tool must adapt to their unique store workflows, not the other way around. Recent updates allow dealers to set custom pricing restrictions, fine-tune routing and transfer rules, and leverage improved multilingual support. This lets the AI adapt to existing store processes rather than forcing new ones. Since February, Call AI has handled more than 8,000 calls in French and Spanish.


Hear the Difference​


Improving Call AI isn’t just about sounding human; it’s about booking showroom visits, retaining shoppers, and protecting team bandwidth. Since February, Call AI has handled nearly 1,900 hours of calls from start to finish without a transfer to staff.


Hear how Call AI handles real test drive bookings, alternative vehicle suggestions, and service rescheduling:


Listen to real Matador Call AI conversations here.


For the dealers here: Where is the biggest phone bottleneck in your store right now? After-hours missed calls, service lane overload, or follow-up speed?
 
  • Like
Reactions: KennySmithAutoUnite
John, I’d say the biggest issue is the handoff after an after-hours sales call.

If somebody calls about a specific VIN, I want the BDC to know which car they asked about, what they were looking for, and who’s responsible for following up when we open. If that unit’s gone, they need to know what alternatives were discussed too.

Otherwise, we’re making the customer start all over again.

Are you tracking how many of those AI-set appointments actually show and turn into sold deals? That’s the number I’d be interested in seeing.
 

✨ AI Highlights

John Kim of Matador AI shares lessons learned from analyzing over 500,000 dealership calls, detailing specific upgrades to their voice AI engine including more natural conversation handling, better management of pauses and interruptions, and improvements aimed at converting more calls into appointments. The post draws on collaboration with a telephony specialist and positions the updates as grounded in real BDC feedback rather than theoretical design. It offers dealership professionals a data-backed look at where AI call handling currently falls short and how targeted fixes can reduce friction in the sales process.

Replies Views 1 104 Started Last Reply