Master Prompts for Dealership Operations: From Customer Quotes to Service Records

Dealership work moves between inventory, pricing, customer communication, financing, service, warranty, and vehicle records. A generic sales chatbot cannot safely own those transitions.

A dealership master prompt should handle one defined communication or documentation task with authorized data, a strict output schema, and explicit review and action permissions.

Customer Quotes

Build the quote from approved inventory, configured vehicle, price, installed products, taxes and fees calculated by the dealership system, offer terms, expiration, and required disclosures.

The model may explain the supplied line items in clear language. It should not calculate a payment, invent a discount, change a fee, promise availability, or present financing terms that were not returned by an authorized system.

{
  "vehicle_id": "inventory-record-id",
  "quote_version": "system-generated-id",
  "verified_line_items": [],
  "required_disclosures": [],
  "draft_message": "",
  "staff_approval": false
}

Lead and Appointment Communication

Use customer consent, channel preference, business hours, staff assignment, vehicle availability, and appointment capacity from controlled systems. Generate a concise response, then let application rules determine whether it may be sent.

Do not infer sensitive traits, pressure a customer, hide terms, or continue messaging after an opt-out. Retain the approved template version and the exact message sent.

Service Intake

Separate the customer's words from the dealership's normalized concern description. Record vehicle identity, mileage, observed symptoms, conditions, prior work supplied by the customer, and requested outcome.

The model can structure an intake draft and identify missing questions. It should not diagnose the vehicle or promise warranty coverage.

Repair Orders and Service Records

Technicians document findings, tests, performed work, parts, measurements, and verification under dealership and manufacturer processes. The prompt may turn approved entries into a customer-facing summary.

Keep technician observations, diagnostic conclusions, recommended work, authorization, declined work, and completed work as different fields. A polished summary must not make recommended work appear completed.

Read AI Prompts for Automotive Technical Documentation for configuration and procedure controls.

Financing, Insurance, and Privacy

Financing, leasing, credit, insurance, identity, and payment data require approved systems, role-based access, applicable notices, and qualified staff. Send only the minimum data needed for the task.

The model should not make credit decisions, infer eligibility, choose a lender or product, or expose application details in logs. Legal and compliance teams define jurisdiction-specific requirements.

A Safe Action Boundary

Model output is data, not authority. Code controls whether a draft can:

• create or change an appointment;
• reserve inventory;
• update a customer record;
• send a message;
• add a repair recommendation;
• request authorization;
• close a repair order.

Every side effect should verify the user, customer scope, current record version, consent, required approval, and idempotency key.

Separate Conversation From the Record

Chat, email, and call summaries are working material. The dealership management, CRM, desking, finance, and service systems hold the approved records. Define exactly which structured fields may be proposed from a conversation and which employee must confirm them.

Preserve direct customer statements where meaning matters. A generated summary should not convert “noise when turning sometimes” into a diagnosis or turn interest in a vehicle into consent for marketing.

When staff correct a name, vehicle, appointment, requested work, or contact preference, write the authorized correction to the system of record and mark the earlier draft superseded. Do not rely on conversation history as the update mechanism.

Review Customer-Facing Language

Test prices, limitations, disclaimers, accessibility, reading level, multiple languages, and channel length. Approved translations and disclosures should be controlled content. Staff should be able to see the source values and edit the draft before it becomes a customer communication.

Test Real Dealership Conflicts

Test sold inventory, changed pricing, expired offers, wrong vehicle identity, a duplicate appointment, unavailable technician capacity, declined work, disputed notes, customer opt-out, unsupported warranty language, and unauthorized access to a finance application.

The system should respond with an exception or handoff, not a convincing workaround.

Staff Own the Relationship

Sales managers own offers and commercial approvals. Finance and insurance staff own regulated product processes. Advisors and technicians own intake, diagnosis, authorization, repair records, and completion. Privacy and compliance teams define data and communication rules. Developers enforce permissions, source freshness, validation, and side effects.

The prompt reduces repeated writing while preserving the people responsible for the customer and vehicle.

Read Master Prompts for Automotive for the broader architecture, then browse dealership workflow contracts in the CyWire marketplace.

This article is technical information, not repair, financing, insurance, regulatory, privacy, or legal advice.