Local & Home Services
AI Phone Agents for Auto Dealerships
Capture every sales lead in under a second, book service appointments around the clock, and stop the $1M+ per store per year bleed from missed calls and hung-up holds.
Dealerships lose more than $1M per store per year to phone failures
The phone is where dealerships quietly bleed out. Service department misses alone cost the average dealership over $1 million per store per year. On the sales side, 56% of new leads come in after hours, but only 37% get addressed within the first hour — and about 70% of people who hit voicemail call a competitor within 30 minutes. Add the 32% of callers who leave voicemails that never get returned and the 31.8% who hang up while on hold, and most dealerships are losing the majority of their phone opportunities before they ever hit the CRM.
The BDC (Business Development Center) exists specifically to fix this, and in the dealerships where it runs well, the results are clear: top-performing BDCs convert 80% of inbound calls to appointments versus the 50% industry average, and responding in under 5 minutes improves closing ratios by up to 900%. But most BDCs are understaffed for peak volume, not staffed at all after hours, and burn through agents quickly enough that quality is inconsistent. The phone problem is not a mystery; it is a staffing problem with a revenue cost most GMs never add up.
AI phone agents fit the dealership phone operation well on both sides of the house. On sales, the AI qualifies inbound lead calls (make, model, budget, trade, timeline, financing), books the appointment with the right salesperson, and runs first-touch follow-up within seconds. On service, the AI handles the highest-volume routine — oil changes, maintenance appointments, recall inquiries, service reminders — which frees the service BDC to focus on complex work orders and warranty questions. Both sides see the speed-to-lead win.
Use cases
Concrete workflows that AI phone agents handle in this industry. Each of these can be wired up with a single phone number, a system prompt, and a set of tools.
- #01
Inbound sales lead qualification
The AI answers every inbound sales call within a second, qualifies the lead against the dealership's criteria (make, model, budget, trade, financing situation, timeline, prior ownership), and books the appointment with the appropriate salesperson or sales manager. Sub-5-minute response, every time.
- #02
Service appointment booking
The highest-volume use case at most dealerships. Customers calling for oil changes, scheduled maintenance, warranty work, recall service, and non-warranty repairs get booked directly into the service schedule without waiting for a service advisor to be free. Covers the 30–50% of service calls that are routine bookings.
- #03
After-hours sales lead capture
Fifty-six percent of new sales leads arrive after hours. The AI handles them at the same quality as business hours — full qualification, appointment booking with the next-available salesperson, and automatic entry into the CRM. Salespeople wake up to booked appointments rather than voicemail lists.
- #04
Service reminder and recall outreach
Outbound AI calls handle service reminders (oil change due, state inspection expiring, OEM maintenance schedule), recall outreach, and recommended service follow-ups. Particularly valuable for recapturing the 46% of 2-year-owners who have stopped coming back to the selling dealer for service.
- #05
Sold customer follow-up
After a vehicle is delivered, an outbound AI call 24–72 hours later checks in on the customer, answers any remaining questions about the vehicle, and nudges for a Google review or CSI survey response. Both satisfaction scores and retention improve when this call actually happens reliably.
- #06
Trade-in inquiry handling
Customers calling to ask about trade-in values get the structured intake (year, make, model, mileage, condition, loan payoff status), a general response from the dealership's valuation knowledge base, and a scheduled appointment for the actual appraisal. Never improvises a specific trade value over the phone.
- #07
Parts department inquiries
Callers looking for parts availability, OEM accessories, or specific part numbers get routed to the parts department directly, or for common items, get the answer from the parts inventory system through a tool call. Frees the parts desk from constant phone interruption.
Auto dealership compliance for AI phone agents
Auto dealerships face a narrower compliance stack than banks or healthcare providers, but the rules that apply are real and actively enforced. The three big areas are TCPA for outbound marketing, FTC Safeguards for customer data, and state-level dealer licensing rules that govern what can be said about pricing, financing, and warranty terms.
TCPA governs outbound calls and texts to consumers. For dealerships, existing customers with an active business relationship can be called about service reminders, recalls, and existing-vehicle communications without separate prior express written consent — but cold marketing calls, campaign outreach to non-customers, and outbound calls about new vehicle promotions require documented consent. Class actions against dealerships for TCPA violations have become common, and the statutory damages ($500–$1,500 per call) add up quickly.
Dealerships are financial institutions for the purposes of the FTC Safeguards Rule because they arrange financing. The 2023 amendments to the Safeguards Rule added specific data-security requirements: encryption of customer information in transit and at rest, access controls, incident response plans, vendor risk assessment, and written information security programs. Any AI phone agent touching customer financial or identifying information must fit into the dealership's existing Safeguards programme.
Every US state licenses auto dealers and regulates what can be stated about pricing, financing, warranties, and vehicle history. An AI phone agent cannot make binding price quotes, guarantee financing approvals, or misrepresent vehicle condition — these are F&I and sales manager decisions that require human accountability. The AI collects the information and books the conversation with a licensed dealer representative.
Dealerships pulling credit reports for financing operate under FCRA, which has specific rules around permissible purpose, disclosure, and adverse action notices. The AI phone agent should not initiate or request credit pulls — that authorisation and the related disclosures belong with the F&I manager, not a voice agent. The AI's job on a financing inquiry is to collect the basic information and book the appointment; the credit application happens with a human.
Important: Dealership compliance is state-specific and the regulators (state DMVs, state attorneys general, FTC, CFPB) actively enforce. Before deploying any outbound AI calling at scale, verify your TCPA consent documentation covers the specific campaigns and the specific entities making the calls, and work with your F&I compliance team to make sure the AI's boundaries are consistent with state advertising rules. BubblyPhone Agents is suitable for inbound lead handling and service appointment booking today; outbound marketing campaigns should be compliance-reviewed before launch.
How to configure an auto dealership AI agent
Dealerships have two distinct phone operations that should usually be configured as two distinct AI agents: a sales BDC agent and a service BDC agent. The sales agent handles inbound leads, qualifies against the dealership's criteria, and books appointments with salespeople. The service agent handles maintenance bookings, recalls, and service reminders, and integrates with the service scheduling system. The two rarely need to talk to each other, and keeping them separate lets each prompt be tuned to its specific audience.
The sales agent's system prompt should define the dealership's lead qualification criteria explicitly. Some dealerships want every lead booked and let the salesperson filter; others want to pre-qualify for budget or financing before booking. The prompt encodes the dealership's actual preference, not a generic 'take every call'. Include the make and model inventory the dealership actually has available — a generic sales agent that cannot answer 'do you have any 2024 F-150s in stock?' loses trust fast.
The service agent's value is in tool integration with the service scheduling system. ServiceDrive, Xtime, myKaarma, and the DMS-native scheduling tools all have APIs that can be exposed as AI tools. The AI queries available slots for the requested service type, books the appointment, and writes the customer record. For dealerships running on older DMS platforms without modern APIs, a thin middleware layer handles the translation — usually a few hours of work for a developer familiar with the DMS.
PATCH /api/v1/phone-numbers/{id}
{
"mode": "streaming",
"system_prompt": "You are the service BDC assistant for Metro Ford. Answer: 'Metro Ford Service, how can I help you today?' Your job is to book service appointments, answer questions about service hours and loaner vehicles, look up whether a specific recall applies to a VIN, and handle maintenance reminders. You do NOT: give mechanical diagnosis, quote exact repair prices without a technician seeing the vehicle, promise warranty coverage without F&I verification, or make any statement about whether a specific problem is covered under warranty. For recall inquiries, collect the VIN or year/make/model and use lookup_recalls. For appointment bookings, collect: customer name, contact, year/make/model of vehicle, the specific concern in the customer's own words, preferred time window, and whether a loaner is needed. Then use book_service_appointment. For customers asking 'how much will it cost' before a technician has seen the vehicle, respond: 'I can get you an accurate estimate once one of our technicians takes a look. Let me get you on the schedule.' Keep the tone warm and efficient — service customers want to be in and out.",
"tools": [
{
"name": "lookup_recalls",
"description": "Check whether any open recalls apply to a specific vehicle",
"parameters": {
"vin_or_description": { "type": "string" }
}
},
{
"name": "check_service_availability",
"description": "Query available service appointment slots",
"parameters": {
"service_type": { "type": "string" },
"preferred_window": { "type": "string" },
"needs_loaner": { "type": "boolean" }
}
},
{
"name": "book_service_appointment",
"description": "Create a service appointment in the dealership's scheduling system",
"parameters": {
"customer_name": { "type": "string" },
"phone": { "type": "string" },
"vehicle": { "type": "string" },
"concern": { "type": "string" },
"scheduled_slot": { "type": "string" },
"needs_loaner": { "type": "boolean" }
}
},
{
"name": "transfer_to_service_advisor",
"description": "Transfer to a live service advisor for complex warranty, diagnostic, or multi-service conversations"
}
],
"tool_webhook_url": "https://your-dealership-api.com/webhooks/tools"
}What it costs compared to alternatives
Auto dealership ROI is dominated by two numbers: the value of a missed service call (often $200–$800 in lost service revenue per incident) and the value of a missed new-car lead (commission-dependent but typically several hundred dollars of front-end and back-end gross for a single conversion). A single captured incremental appointment per day pays for the AI agent many times over.
Scenario: A single-rooftop franchise dealership handling 4,000 calls per month across sales and service (average 3 minutes per call).
| Option | Cost | Notes |
|---|---|---|
| Existing BDC only (baseline) | $8,000 – $20,000 / month | Fully loaded BDC with 2–4 agents covering sales and service. Misses after-hours, peak volume overflow, and has variable quality depending on staffing. Top BDCs convert 80% of inbound calls; most convert closer to 50%. |
| Outsourced BDC / answering service | $2,000 – $6,000 / month | Third-party BDC handles overflow and after-hours. Quality varies significantly by vendor. Cannot always integrate with the dealership's DMS or scheduling tools. |
| BubblyPhone Agents (service BDC only) | ~$950 / month | 12,000 minutes × $0.04/min + $0.04/min model + $3/mo number. Picks up every service call instantly, books directly into the service schedule via tools, handles recall lookups and reminders. |
| BubblyPhone Agents (sales + service, both lines) | ~$950 – $1,900 / month | Same per-minute rate across both phone numbers. Runs sales lead qualification and service booking as two separate agents sharing the account-level billing. |
For dealerships already spending $8K–$20K/month on a BDC, the AI agent is not a replacement for the team — it is an overflow and after-hours layer that recovers the 50% of calls the team misses. At $1 million per year in lost service revenue, even modest recovery rates pay for the system many times over.
Frequently asked questions
Does the AI integrate with my DMS?
Through tool calling, the AI can integrate with any DMS that has an API — CDK Global, Reynolds & Reynolds, Dealertrack, Auto/Mate, DealerCenter, and similar. The AI calls the DMS through a tool during the call, queries available appointment slots, writes customer records, and books the appointment. For legacy DMS platforms without modern API access, a thin middleware layer handles the translation, usually a few hours of work. Most dealerships find the service scheduling integration easier than the CRM integration because Xtime, myKaarma, and ServiceDrive have well-documented APIs.
Can the AI quote a vehicle price or trade-in value?
It should not. Price and trade value negotiation is F&I and sales management territory, and the AI committing to a specific number over the phone creates pricing exposure and customer expectations the dealership may not be able to honour. The AI's job is to collect the information a salesperson or appraiser needs, give ranges from the public inventory when appropriate, and book the in-person or structured appointment where the actual numbers get discussed. Most dealerships find this boundary sharpens their process rather than limiting it.
How does the AI handle service pricing questions?
By ranges, not commitments. The AI reads from a service pricing knowledge base with typical ranges for common work (oil change, tire rotation, brake pads, state inspection) but never commits to an exact price without a technician on the vehicle. For diagnostic requests ('what will it cost to fix my transmission'), the AI says a technician needs to look and books the diagnostic appointment. This is the same discipline the best human service advisors use.
Can the AI replace the BDC entirely?
For most dealerships, no — and probably not. The AI is an overflow and after-hours layer that captures the calls the human BDC cannot get to, plus the routine bookings that tie up BDC bandwidth. Complex conversations (finance scenarios, trade negotiations, sales manager escalations, warranty disputes) stay with humans. Dealerships that try to fully eliminate the BDC tend to find that the 20% of calls the AI cannot handle well are the ones that drive the most revenue.
How does the AI handle TCPA compliance for outbound service reminders?
Existing customers with an active business relationship — meaning customers who purchased or had service performed at the dealership — can be called about service reminders, recall notices, and existing-vehicle communications under TCPA's established business relationship exception. The relationship has a time-based limit (typically 18 months since last transaction, depending on jurisdiction), so older customers whose relationship has lapsed need documented consent to call. For cold prospecting or marketing to non-customers, you need documented prior express written consent — same as any other outbound marketing. Work with your compliance team to scrub lists before any campaign.
What about Spanish-speaking customers?
Both streaming-mode models BubblyPhone supports (Gemini Live and GPT Realtime) handle Spanish natively, along with most other major languages. The system prompt should instruct the AI to detect the caller's language and respond in the same one. For dealerships in markets with significant Spanish-speaking customer bases, this is one of the easiest wins — the same AI handles English and Spanish callers with equal fluency, with no separate staffing or IVR branching.
Build a local & home services AI phone agent today
Purchase a number, wire up your tools, and have a working agent answering real calls by the end of the afternoon.