Best AI in the dealership, what’s useful / something wish you had?

[jessh]

Hey everyone! I'm pretty new to this forum but really interested in discussions new tech / AI in the dealership world.

I feel like I see 10+ companies all chasing the same problem in the software world with slightly different spins. In the dealership world, what are the actual use cases you’re seeing AI help with today, or ones you wish existed?

And is bringing AI a top priority or just something nice to have?

 

[kentmakishima]

Welcome to the forum!

I agree in the software world there is a concentration around the low hanging applications (chat, copy writing, etc) but the market is big enough for many players to get a slice. In the dealer world, we see similar AI-forward applications in chat and marketing. I expect to see it integrated behind the scenes to existing service providers for CRMs, DMS, and other data-heavy platforms.

For my company, AI is integral in our data infrastructure that powers our products for our dealer customers (ex: vehicle acquisition across FB marketplace, forums, marketplaces, etc)

 

[jessh]

Welcome to the forum!

I agree in the software world there is a concentration around the low hanging applications (chat, copy writing, etc) but the market is big enough for many players to get a slice. In the dealer world, we see similar AI-forward applications in chat and marketing. I expect to see it integrated behind the scenes to existing service providers for CRMs, DMS, and other data-heavy platforms.

For my company, AI is integral in our data infrastructure that powers our products for our dealer customers (ex: vehicle acquisition across FB marketplace, forums, marketplaces, etc)

Thanks Kent! Interesting point on the data infrastructure side - I can see where it plays into Hypercars (cool platform!). Definitely curious on the behind the scenes integrations with existing providers, whether there's an angle of niche AI training of human-in-the-loop workflow tools.

 

[Steve Roessler]

Hey Jess, there are many functions of AI utilization right now. Any particular area you are asking for specifically?

Based on your last reply, AI is helping minimize workflows for people. Many stores are decreasing work flows from 90 or 120 days down to less than 10 days. Instead of faking activities, AI will go out there and create engagement and tee it up for a human. You just have to ask your AI provider to provide you the "triggers" on what AI will do and re-design your workflow. Hope that makes sense.

 

[althompson707]

Great question, Jess - and you nailed the core frustration right in your opener.
You’re right that there are 10+ companies chasing the same problem with slightly different spins. The honest answer is: most of them are wrappers. They’re stitching together ChatGPT, a CRM API, and a voice vendor and calling it an AI BDC platform. That matters a lot when you’re evaluating what actually works in a dealership.
What’s genuinely useful today - use cases I see working:
• AI BDC agents for automotive handling inbound leads 24/7 across SMS, email, chat, and voice - without a human touching it until the appointment is set. Not a chatbot. An actual agentic system that reasons, responds in context, and follows up.
• Voice AI agents for car dealers that handle after-hours calls, service reminders, and internet leads at a quality level that’s hard to distinguish from a trained BDC rep - and at a fraction of the cost.
• AI-driven lead follow-up that personalizes outreach based on what the shopper actually did on the website - what inventory they viewed, how long they spent, what they configured. Most dealers are still blasting generic drip sequences. That era is over.
• Conversational AI for dealerships that actually updates the CRM, routes to the right salesperson, and retains context across every channel - so the customer never has to repeat themselves.
What doesn’t work (yet): Anything bolted onto a broken process. AI amplifies what’s already there - good or bad.
The thing most dealers don’t realize they’re missing:
The real moat isn’t the AI model. It’s the data layer underneath it. The best automotive AI platforms are building what I’d call a self-healing customer intelligence graph - a unified picture of every customer signal across web, SMS, email, phone, and chat - that gets smarter with every engagement. That’s what separates a platform from a point solution.
OEM certification matters too. If your AI is sending non-compliant first-response emails or violating brand standards on voice, you have a compliance problem. There are very few AI platforms for auto dealers that have gone through OEM certification. That alone filters out most of the field.
On your question - is AI a top priority or nice to have?
For any dealer doing $5M+ gross a year, the math on AI BDC vs. human BDC headcount is so compelling right now that it’s quickly becoming a competitive disadvantage not to adopt. Dealers running AI voice agents and agentic follow-up are seeing response times under 60 seconds, 24/7, with engagement rates that outperform their human BDC teams in the evening/weekend windows where 40% of internet leads come in.

Full disclosure: I’m Albert Thompson, CEO of ID Privacy AI - a full-stack agentic AI platform built specifically for automotive retail. We’re OEM certified with Nissan, Infiniti, and Mitsubishi, and we’ve run 1M+ agentic engagements across 150+ dealers. I tried to give you the honest lay of the land above regardless - happy to answer any questions or geek out on the tech side of this.

 

[Auntess]

Great question, Jess - and you nailed the core frustration right in your opener.
You’re right that there are 10+ companies chasing the same problem with slightly different spins. The honest answer is: most of them are wrappers. They’re stitching together ChatGPT, a CRM API, and a voice vendor and calling it an AI BDC platform. That matters a lot when you’re evaluating what actually works in a dealership.
What’s genuinely useful today - use cases I see working:
• AI BDC agents for automotive handling inbound leads 24/7 across SMS, email, chat, and voice - without a human touching it until the appointment is set. Not a chatbot. An actual agentic system that reasons, responds in context, and follows up.
• Voice AI agents for car dealers that handle after-hours calls, service reminders, and internet leads at a quality level that’s hard to distinguish from a trained BDC rep - and at a fraction of the cost.
• AI-driven lead follow-up that personalizes outreach based on what the shopper actually did on the website - what inventory they viewed, how long they spent, what they configured. Most dealers are still blasting generic drip sequences. That era is over.
• Conversational AI for dealerships that actually updates the CRM, routes to the right salesperson, and retains context across every channel - so the customer never has to repeat themselves.
What doesn’t work (yet): Anything bolted onto a broken process. AI amplifies what’s already there - good or bad.
The thing most dealers don’t realize they’re missing:
The real moat isn’t the AI model. It’s the data layer underneath it. The best automotive AI platforms are building what I’d call a self-healing customer intelligence graph - a unified picture of every customer signal across web, SMS, email, phone, and chat - that gets smarter with every engagement. That’s what separates a platform from a point solution.
OEM certification matters too. If your AI is sending non-compliant first-response emails or violating brand standards on voice, you have a compliance problem. There are very few AI platforms for auto dealers that have gone through OEM certification. That alone filters out most of the field.
On your question - is AI a top priority or nice to have?
For any dealer doing $5M+ gross a year, the math on AI BDC vs. human BDC headcount is so compelling right now that it’s quickly becoming a competitive disadvantage not to adopt. Dealers running AI voice agents and agentic follow-up are seeing response times under 60 seconds, 24/7, with engagement rates that outperform their human BDC teams in the evening/weekend windows where 40% of internet leads come in.


For teams that are actively testing different AI workflows and comparing how these systems actually behave in practice, tools like Atomic Chat are sometimes used to quickly benchmark responses and reasoning quality across different models and setups before committing them into production workflows.




Full disclosure: I’m Albert Thompson, CEO of ID Privacy AI - a full-stack agentic AI platform built specifically for automotive retail. We’re OEM certified with Nissan, Infiniti, and Mitsubishi, and we’ve run 1M+ agentic engagements across 150+ dealers. I tried to give you the honest lay of the land above regardless - happy to answer any questions or geek out on the tech side of this.

It's a good review, but it does contain a few points that, in practice in the dealer environment, often appear slightly less "perfect" than they do in the description.

It's hard to argue with the fact that most solutions today are simply wrappers over LLM + CRM + voice layer. This is true, and the problem isn't the use of ChatGPT or similar solutions, but the fact that a proper product layer is often missing: routing logic, data quality control, and stable integration with DMS/CRM. As a result, "agency" ends where real operational complexity begins.

 

[Carl Schiller]

Great question, Jess - and you nailed the core frustration right in your opener.
You’re right that there are 10+ companies chasing the same problem with slightly different spins. The honest answer is: most of them are wrappers. They’re stitching together ChatGPT, a CRM API, and a voice vendor and calling it an AI BDC platform. That matters a lot when you’re evaluating what actually works in a dealership.
What’s genuinely useful today - use cases I see working:
• AI BDC agents for automotive handling inbound leads 24/7 across SMS, email, chat, and voice - without a human touching it until the appointment is set. Not a chatbot. An actual agentic system that reasons, responds in context, and follows up.
• Voice AI agents for car dealers that handle after-hours calls, service reminders, and internet leads at a quality level that’s hard to distinguish from a trained BDC rep - and at a fraction of the cost.
• AI-driven lead follow-up that personalizes outreach based on what the shopper actually did on the website - what inventory they viewed, how long they spent, what they configured. Most dealers are still blasting generic drip sequences. That era is over.
• Conversational AI for dealerships that actually updates the CRM, routes to the right salesperson, and retains context across every channel - so the customer never has to repeat themselves.
What doesn’t work (yet): Anything bolted onto a broken process. AI amplifies what’s already there - good or bad.
The thing most dealers don’t realize they’re missing:
The real moat isn’t the AI model. It’s the data layer underneath it. The best automotive AI platforms are building what I’d call a self-healing customer intelligence graph - a unified picture of every customer signal across web, SMS, email, phone, and chat - that gets smarter with every engagement. That’s what separates a platform from a point solution.
OEM certification matters too. If your AI is sending non-compliant first-response emails or violating brand standards on voice, you have a compliance problem. There are very few AI platforms for auto dealers that have gone through OEM certification. That alone filters out most of the field.
On your question - is AI a top priority or nice to have?
For any dealer doing $5M+ gross a year, the math on AI BDC vs. human BDC headcount is so compelling right now that it’s quickly becoming a competitive disadvantage not to adopt. Dealers running AI voice agents and agentic follow-up are seeing response times under 60 seconds, 24/7, with engagement rates that outperform their human BDC teams in the evening/weekend windows where 40% of internet leads come in.

Full disclosure: I’m Albert Thompson, CEO of ID Privacy AI - a full-stack agentic AI platform built specifically for automotive retail. We’re OEM certified with Nissan, Infiniti, and Mitsubishi, and we’ve run 1M+ agentic engagements across 150+ dealers. I tried to give you the honest lay of the land above regardless - happy to answer any questions or geek out on the tech side of this.

Albert covered the sales and BDC side of this really well, and he's right that the data layer is what separates a real platform from a wrapper. I'd add one area that this thread hasn't touched yet: the service lane.


The fixed ops side of the dealership is where I spend most of my time, and it's genuinely the last room that purpose-built AI hasn't reached. Chat, BDC, lead follow-up all have real solutions now. But the service drive check-in is still almost entirely manual. An advisor talks to a customer for 10 minutes, tries to hold the vehicle history in his head, cross-references what he remembers about open recalls, and writes up a repair order based on what he caught. On a busy Monday morning with 40 cars on the board, that's where revenue walks out the door.


The use case I wish had existed two years ago (and what we're building now) is an AI layer that listens to that check-in conversation in real time, cross-references VIN-level service history and OEM repair data, and surfaces structured recommendations to the advisor before the customer walks to the waiting room. Not replacing the advisor. Just making sure nothing gets missed.

I'm Carl Schiller, co-founder of MichigAIn Automotive Solutions. We're about to enter the pilot phase at a Michigan GM rooftop and building from there. Happy to compare notes with anyone else working the fixed ops side!

 

[realmatkoenig]

I second what @Steve Roessler said and I'll add that for our clients AI has been helping them by doing the work the salespeople typically aren't consistently strong at. Appointment setting, follow up, etc,. Most of our dealers use AI to handle the initial lead then pass it off to a salesperson when the appointment is set (or alert them if it can't set one). It allows their sales team to focus on things that require a human touch. Hope that helps

Hey Jess, there are many functions of AI utilization right now. Any particular area you are asking for specifically?

Based on your last reply, AI is helping minimize workflows for people. Many stores are decreasing work flows from 90 or 120 days down to less than 10 days. Instead of faking activities, AI will go out there and create engagement and tee it up for a human. You just have to ask your AI provider to provide you the "triggers" on what AI will do and re-design your workflow. Hope that makes sense.

 

[Lotsmith]

Lot of sharp takes here on the BDC, voice, and CRM side. The gap I don't see mentioned much is the marketing side of the lot: not lead handling, but getting every car in front of buyers in the first place.

Where I keep seeing AI actually earn its keep there is the unglamorous part, marketing the whole lot instead of just the fresh trades. Most stores get the new arrivals shot and listed, and the aged units quietly pile up because nobody has the hours to re-photograph and rewrite every one. AI that reads your inventory, flags what's aging, and drafts the listing (photos and copy built around who the car is actually for) saves real time, as long as a person still approves it before it goes out.

Full disclosure, I'm building in this exact space (Lotsmith), so grain of salt. But even setting that aside, it comes back to the point a few of you made in here: does it actually work off your real inventory and hold up operationally, or is it a generic writer you still feed and post car by car? Happy to trade notes.

 

[dealerdata1]

I have yet to see a solid AI implementation from the legacy vendors

And anything to do with lead handling or response and AI is a complete no-go in my book

If a customer is actively reaching out to your business, a real human should be responding. I compare it to calling your insurance provider and getting into an argument with the AI operator and you eventually just start yelling "SPEAK TO A REPRESENTATIVE" or some variation

If I was in store today, I would look at every administrative task my staff does on a per department level and ask:
Can this be automated without AI?
If it can be automated, what use would AI have in assisting the automation?

So if it were me, deal jackets. Automating deal jacket processing without any AI is so incredibly easy to make a workflow in Zapier or some other similar provider. You don't even need AI to pre-built deal jackets, check for spelling errors or mandatory blank fields, automated routing or notifications when certain conditions are met. The list goes on with the possibilities you could do with deal jackets alone.