

Every vendor at every dealer conference for the last two years has promised the same thing: an AI that answers your phone, doesn't get tired, and never loses a lead to voicemail. Most of that has been slideware. Cars24 — India's largest used-car marketplace — just published enough hard numbers to make the pitch worth a second look.
The company deployed OpenAI-powered voice and chat agents across its customer service operation and is now running over 1 million conversation minutes a month through them. That's not a pilot. That's a phone system.
Strip out the marketing language and three numbers hold up under scrutiny:
That last figure is the one that should matter most to a dealer operator, because it's not a productivity metric — it's revenue that was already gone. Every store has a graveyard of leads that went cold because nobody called back in time, or called back with the wrong information, or called back at all. Recovering even a low double-digit percentage of that pile is real money, and it didn't require a human to do it.
Worth noting: this isn't a chatbot with a script. Cars24's system is agentic — it gathers vehicle details, books appraisal appointments, sends pre-appointment reminders, and automatically reschedules when a customer doesn't show. It hands off to a human sales rep only when the conversation actually needs one. Voice-handled conversations reportedly grew 30% quarter over quarter, which tells you customers aren't just tolerating the bot — they're increasingly comfortable talking to it instead of typing.
The reason this case study matters beyond one used-car marketplace in India is the timing. It landed in the same week as three other moves that all point the same direction:
DriveCentric shipped a "Service-to-Sales Agent" that mines CRM, vehicle, and service history data to flag customers in for an oil change who are statistically ready to trade in — then starts that conversation itself before handing off to a salesperson. Early access is live now, full release August 1.
BMW put its vehicle configurator directly inside ChatGPT, letting shoppers spec a car through conversation rather than a web form. That moves the point of sale outside the dealer's own website entirely and into a chat window the customer already has open.
Sable, an AI "sales rep" that operates web pages and demos products live on a shared browser, raised $45 million from Sequoia and 8VC — evidence that investors think AI handling actual sales conversations, not just answering FAQs, is where the money goes next.
Put together, the pattern is clear: AI agents are moving from "answer the phone" to "run the funnel." Cars24 is the first with hard numbers proving the phone-answering part works. DriveCentric and BMW are betting the next layer — trade-in triggers and mid-conversation product configuration — works too.
Don't buy a voicebot because Cars24 published a good number. Buy one if you can answer three questions first: Where are your leads actually dying — after-hours calls, slow follow-up, or missed voicemail? Does your CRM have clean enough data to let an agent act on it without creating a mess a human has to clean up later? And can you measure recovered leads specifically, the way Cars24 did, rather than vague "engagement" metrics that vendors love and mean nothing?
The Cars24 numbers are a legitimate proof point — a 12% lead recovery rate, at volume, is not a rounding error. But it's proof for a used-car marketplace running over a million minutes a month with an engineering team behind it, not proof that any dealership plugging in a chatbot gets the same result. The gap between those two things is where most AI vendor pitches quietly fall apart. Ask for the number before you sign.

Cars24 deployed OpenAI-powered voice and chat agents across its used-car buying and selling funnel, and the numbers are unusually specific for a vendor case study: 50% faster support resolution, 80% less staff time on appraisal tasks, and 12% of abandoned sellers pulled back into the pipeline. It's one of the more credible AI deployment stories in retail auto this year — real conversation volume, real recovered deals, not just a demo reel.

2026 survey data shows dealers are pouring budget into AI, with voice agents and speed-to-lead tools topping priority lists — and generating measurable ROI. But most stores are still stuck on surface-level chatbots while GM's bet on Impel signals where the real infrastructure fight is heading: an integrated AI operating layer, not another point solution.