It's 11 PM on a Thursday in Riyadh. A customer is browsing the MINI website on his phone. He's interested in the Countryman, wants to know if it comes in Island Blue, and has a question about the monthly payment. Nobody's there. The showroom closed four hours ago. The WhatsApp number goes to a phone sitting on an empty desk.
By Friday morning, he's put down a deposit at another dealer.
This happens hundreds of times a week across the Middle East. The cars are there. The demand is there. The staff is not.
The dealer
Mohamed Yousuf Naghi Motors is one of the largest automotive groups in the Middle East. BMW and MINI across Saudi Arabia. Dozens of showrooms. Thousands of vehicles in inventory. A sales operation that runs on people — people who work fixed hours in a market that shops around the clock.
The challenges were specific and measurable. After-hours inquiries went unanswered. Arabic-speaking customers got English-only responses, or no response at all. Service bookings — oil changes, tire rotations, brake inspections — consumed hours of staff time per day on phone calls that followed the same script every time. And leads walked away to competitors who happened to pick up the phone.
What we deployed
Salesteq deployed an AI Sales Advisor across the MINI dealership network. Not a chatbot with a decision tree. Not a FAQ page with a search bar. An AI that knows the entire vehicle inventory — every model, every trim, every color, every price — and can hold a real conversation about it.
The advisor runs on three channels: the dealership website, phone (via Twilio), and WhatsApp. It speaks Gulf Arabic and English natively, switching between them mid-conversation if the customer does. It handles everything from product questions to test drive bookings to service appointments.
The underlying system connects to live inventory data through Yeda, our knowledge layer. When a customer asks "Do you have the Countryman in blue?", the advisor doesn't guess. It queries the actual inventory, returns the specific units available, with pricing and specs.
How it actually works
A customer messages at 11 PM asking about the MINI Countryman in Arabic. The advisor responds in Arabic. It knows the Countryman S starts at SAR 169,000, comes in 12 colors, and has a 2.0L turbocharged engine with 204 hp. The customer asks how it compares to the Clubman. The advisor explains the differences — cargo space, wheelbase, ride height — without hedging or filler.
The customer wants a test drive. The advisor checks available slots, confirms a time, captures the customer's name and phone number, and books it. Done. No form. No "someone will get back to you." No waiting until Sunday when the showroom reopens.
Another customer calls the dealership number at 7 AM asking for an oil change appointment. The voice advisor picks up, confirms the vehicle model and year, checks the service calendar, and books the appointment in 30 seconds. The entire interaction takes less time than it would take a human to pull up the booking system.
Service bookings were the surprise win. The volume of repetitive service calls — "I need an oil change," "when can I bring my car in," "how much is a brake inspection" — was consuming significant staff time every day. These are conversations that follow an identical pattern every time. The AI handles them faster and more accurately than a human on the phone, because it never puts anyone on hold and never has to look anything up.
The numbers
The advisor runs 24 hours a day, 7 days a week, across every channel simultaneously. It handles Arabic and English without a language toggle or a separate bot. It books test drives and service appointments without human intervention.
There was no new headcount. No night shift. No additional call center seats. The same team runs the same operation, except now every inquiry gets answered in under two seconds, in the customer's language, at any hour.
The staff that used to spend hours on service booking calls now spend that time on high-value activities — closing deals, building relationships, handling complex trade-ins. The AI didn't replace them. It took the repetitive work off their plate.
This is running right now
This isn't a pilot. It's not a proof of concept. It's not a slide deck about what AI could do for automotive retail someday. It's production software handling real customer conversations at one of the largest dealer groups in the Middle East.
You can try it yourself. The MINI KSA sales advisor is live at mini-ksa.com/chat-test. Ask it about a car. Book a test drive. Switch to Arabic mid-sentence. It works.
Every automotive dealer has the same problem Naghi had: customers who shop outside business hours, service calls that eat up the day, and leads that disappear because nobody answered fast enough. The technology to fix it exists today. It's not coming next year. It's running right now.