AI Command Center is generally available today – after seven months, 119 dealerships, and thousands of real dealer questions put it through its paces. Here’s why we built it that way.
We Held Back Our AI for Seven Months. On Purpose.
AI Command Center is generally available today – after seven months, 119 dealerships and thousands of real dealer questions put it through its paces. Here’s why we built it that way.
Today, Lightspeed AI Command Center™ is generally available.
There’s no shortage of AI announcements right now. So rather than give you another list of AI features, I want to explain why we waited seven months to make this one.
We had a working version of Lightspeed AI Command Center in January.
We could have launched it then.
We didn’t.
Instead, we put it in the hands of dealers and asked them to test it, challenge it, find where it fell short and help us make it better. We decided we wouldn’t release AI Command Center until the people who would actually use it trusted the answers it gave them.
That decision shaped the product we’re launching today.
The problem isn’t a lack of data
Over the past year, our product team has spent a lot more time where our customers actually work: inside dealerships. We now average 10–12 dealer visits every month.
Across dealerships of every size and every industry we serve, we kept hearing versions of the same problem:
Dealers have the data they need to make better decisions. Getting to it is the hard part.
Revenue. Gross profit. Unit sales. Inventory aging. Salesperson performance. Market penetration. Service performance.
The information exists. But too often, finding an answer means running reports, exporting spreadsheets, searching across systems or asking someone else to pull the numbers.
By the time the answer arrives, the opportunity to act may already be disappearing.
We thought there should be a better way.
Stop Guessing. Just ask.
Imagine asking:
- Which units are aging?
- Where is gross profit slipping?
- How is my sales team performing?
- Where am I underpenetrated in my market?
And getting an immediate answer grounded in your dealership’s live data.
Not just a number, either.
AI Command Center can give you trend lines, charts, graphs, comparisons and breakdowns by store, brand, salesperson or month. Then you can ask a follow-up questions and keep digging deeper through conversation.
The goal isn’t to make dealers better at reporting. It’s to make reporting less necessary.
As we said in announcing AI Command Center, dealers shouldn’t have to spend their time searching for information. They should be acting on it.
That’s the opportunity we saw in AI: move dealers from question to action while the opportunity still exists.
Lightspeed has spent more than 40 years building technology for dealerships and today supports more than 4,500 dealers. Across that history, Lightspeed’s data represents more than $159 billion in dealership transactions, 1.7 million units sold and 12.8 million service repair orders.
Combine that industry context with each dealership’s own live operational data, and AI can begin to understand not just numbers, but what those numbers mean inside a dealership.
We didn’t want to launch another generic AI tool
AI Command Center isn’t a consumer AI tool pointed at dealership data from the outside.
It lives inside Lightspeed DMS.
There’s no separate analytics application to learn. No exporting dealership data into another platform. No duplicate database to maintain. No new permissions model to administer.
A manager asks a question inside the system they already use, in the same kind of language they’d use to ask a colleague.
And because existing Lightspeed permissions remain intact, employees only receive answers based on information they’re already authorized to access.
That distinction matters.
The true power of AI isn’t generating more content. It’s helping people make better decisions.
For us, that meant AI needed to become a natural extension of the DMS – not another destination dealers had to visit.
Built for dealers. By dealers.
This is where the story gets interesting.
We opened an AI Pioneers program for dealers willing to use an early version of AI Command Center and help us shape it.
We expected interest.
We didn’t expect 119 dealerships to sign on.
1. The AI Pioneers program
We opened an AI Pioneers program to early-adopter dealers who were willing to test the product and shape it for our industry. 119 dealerships signed on — exceeding our expectations. They tested it, molded it, told us where it fell short, and added items to our roadmap that we would not have thought of on our own. I can’t thank them enough for their help. Here’s a breakdown of their demographics:
By role. Two out of every five Pioneers are owners or dealer principals. Another quarter are General Managers.
- 42% Owners and Dealer Principals
- 24% General Managers
- 12% Controllers, CFOs, and office managers
- 7% VPs and corporate executives
- 6% IT and systems leads
- The remainder span across sales, service, parts, operations, and marketing leadership
By industry. Every segment we serve is represented, and more than a third of Pioneer dealerships carry units in multiple industries – which is exactly the complexity we needed the product tested against.
- 55% Powersports
- 53% Marine
- 23% Trailer
- 17% RV
- 9% Golf
- 37% of Pioneer dealerships operate in more than one industry, so these figures add to more than 100%
By size. From single-rooftop independents to multi-store groups with dozens of locations.
2. A proprietary AI Trust Score
This is a part we’re excited about, and something most of the market is skipping. Most AI products launch when the demo looks good. We decided AI Command Center wouldn’t launch until we could measure that dealers actually trust the answers. So, we built the Lightspeed AI Trust Score™. Every question a dealer asks is captured and can be scored across five dimensions, each on a 0-to-5 scale:
- Data Accuracy: Are the numbers right, and do they reconcile to what’s actually in the DMS?
- Answer Relevance: Did it answer the question the dealer actually asked?
- Groundedness: Is every claim traceable to the dealer’s own data, with no invention?
- Response Quality: Is the answer clear, complete, and usable by a non-technical person?
- Visualization Quality: Is the chart or table the right one, and does it read correctly?
Those five roll up into a single AI Trust Score. Scoring comes from three places: direct dealer feedback in the product, hands-on review by our team through real dealer sessions, and an AI agent that runs nightly across every session to evaluate output quality and flag anything that needs a human look.
We decided we would not release AI Command Center until our frequent users scored it consistently at 4.0 out of 5 or better, averaged across all five dimensions. Early on, we weren’t close, AI Pioneer testing surfaced real problems that we would never have found in a lab. We fixed them, and re-tested with the same dealers. The reason why I am writing this post in August is because we have hit our 4.0 AI Trust Score threshold.
We didn’t launch when we thought AI Command Center was ready. We launched when dealers told us it was.
What that means for dealers starting today
If you’re adopting AI Command Center now, you’re not starting with the version we had in January.
You’re starting after 119 dealerships have already pushed it through thousands of real-world questions.
Many of the obvious gaps have already been found.
Questions that produced answers that sounded right but weren’t have been identified and addressed.
And the system has been tested against the messy reality of dealership operations – not a perfectly curated demo environment.
That’s a critical difference.
It knows dealerships. Then it learns yours.
Purpose-built AI also needs context.
There is a real difference between AI that is part of your dealer management system, and AI that is bolted on from the outside. (This is where you should look closely at anything you’re being sold right now.)
AI Command Center lives inside Lightspeed natively. It is NOT an integration.
AI Command Center is a platform built into Lightspeed DMS — not a third-party product connected by an integration, and not a separate app you log into. You open it from inside the DMS you already use. There is no data hand-off to another vendor, no additional contract with a company you’ve never heard of, and no new user roles, permissions or security protocols to set up.
It runs on your data — and only your secure data.
Every answer is generated against your dealership’s own live operational data. Not a cached spec sheet. Not a public data source. Not a blend of the two that you can’t untangle. When AI Command Center gives you a margin number, that number came out of your DMS. And your data is yours. Per our data privacy agreements, it is not shared, sold, or used to train external models. Your secret sauce — how you price, how you stock, how you pay — stays inside your DMS and business.
AI Command Center honors Lightspeed’s existing role-based access and field-level security. A user can only ask about data they already have the right to see. You don’t have to administer a second permissions model, and you don’t have to worry about an employee asking a question that returns payroll if they don’t have access to it in the first place.
It learns how you run your store — we call this your Dealer Blueprint.
Understanding the context of your business is where most AI products fall down. Two dealers can ask the identical questions and need different answers, because they don’t define certain words, calculate units or margin, or run their business the same way.
AI Command Center is built on two layers of data context. Dealer DNA teaches the system how our industries actually work — what a unit is, how a floorplan works, what F&I means, how seasonality hits a marine store differently than a powersports store. Your Dealer Blueprint goes a layer deeper and teaches it how — specifically — your dealership runs.
Here’s what that difference looks like in practice. Take a question a dealer user could ask like: “I have a 2024 unit that has been sitting on the floor for 95 days, what should I do?”
Without a Blueprint, a generic AI may recommend something reasonable but very wrong if it is looking at public Internet data. It could say something like “Drop the price 15%,” when dealer leadership does not want to immediately fall back on discounting prices.
With your Blueprint, it already knows your flooring, discounting and pricing rules. This answer would come back with a better answer for this partially dealer, something like, “Per our pricing rules, hold list. Run a 30-day promo to pre-qualified leads. Obtain financing approval if any discount exceeds 10%.”
Apply this to how you calculate margin, when you close your books, how you manage floor plans, etc. This is data and context you control. Your Blueprint lives in your system settings and AI Command Center references it before it gives answers on your business.
That moves AI from generic advice toward something much more valuable: an answer grounded in the way your business actually operates.
And your dealership data stays your dealership data
There’s another reason we wanted AI inside Lightspeed.
Dealers shouldn’t have to choose between getting useful AI insights and protecting sensitive business information.
AI Command Center operates directly within the Lightspeed environment. Existing role-based permissions and field-level security continue to apply.
If an employee isn’t authorized to see certain information in Lightspeed, AI Command Center doesn’t suddenly give them access to it.
And dealership data isn’t shared, sold or used to train external models.
Your pricing strategy, inventory, financial information and operating practices remain yours.
What does this look like in the real world?
One AI Pioneer dealer had spent years trying to understand market penetration across the two counties where he sold most of his units.
He knew the markets felt different. He just couldn’t get to a reliable answer.
So he asked AI Command Center: What’s my market penetration by county?
AI Command Center analyzed his sales history against market registration data.
The answer surprised him.
His market share was 54% in one county – and only 15% in the other.
Suddenly, his marketing strategy looked very different. Instead of continuing to spend heavily in the county he already dominated, he could focus more resources on the market where there was considerably more room to grow.
As another AI Pioneer, Matt Weller of Weller Recreation, described his experience, he went into the program looking for gaps in his OEM lineup and came away realizing the larger opportunity wasn’t adding brands – it was increasing market share inside his existing PMA.
That’s what we’re trying to unlock.
Not more data.
Decisions you couldn’t see before.
Sometimes the win is simply getting half a day back
A GM of a four-rooftop group had already figured out how valuable AI could be.
His process was to export Lightspeed data, clean it in Excel, load it into a consumer AI tool and keep asking questions until he reached an answer.
It worked.
It also took about half a day every time he wanted to investigate something new – and required putting dealership financial information into an outside AI platform.
That’s the hidden cost of a workaround.
It isn’t only the hours.
It’s the questions you stop asking because getting the answer takes too much time or creates too much risk.
With AI Command Center, that workflow becomes a conversation inside Lightspeed.
No export. No cleanup. No separate AI tool.
Just ask.
We kept the pricing simple, too
Our AI Pioneers were clear about this one: don’t make AI pricing another thing dealers have to decipher. AI is new enough without token-based usage models. And dealers who’ve used AI know that token costs climb with usage – which means unpredictable pricing is a real risk. .
So our first offering is a simple flat monthly fee that ensures predictability and minimizes risk. That fee includes unlimited user access to AI Command Center and the first data set it is running on, your dealer sales data. More data sets and features are coming on the roadmap— service, parts, accounting, and beyond.
Over time, as we add new AI capabilities, we may introduce new tiers — an entry tier and a more advanced one. But the first release of AI Command Center is a flat monthly fee, with data sets and core features included in that price.
This is the beginning, not the finish line
The initial release of AI Command Center focuses on conversational intelligence around sales performance and inventory management.
But the vision is much larger.
Over time, we see conversational intelligence extending to more dealership roles and departments, creating a unified AI experience across the business.
The idea is simple:
Every dealership employee should eventually be able to ask a business question in plain English and receive an immediate, trustworthy answer based on the information they’re authorized to access.
That represents a fundamental shift in dealership software.
From searching for data to having a conversation with it.
From reports to answers.
From answers to action.
Seven months later, here’s what matters
If you remember four things about AI Command Center, make them these:
- It’s tested. Seven months. 119 dealerships. Thousands of real questions before general availability.
- It’s measured. We created an AI Trust Score and required the product to reach 4.0 out of 5 with frequent users before we released it.
- It’s yours. Your live dealership data, your permissions, inside Lightspeed.
- It’s simple. Ask a question in plain English and get an answer you can act on.
There’s a lot of noise around AI right now.
We don’t think dealers need more AI noise.
They need a faster, easier way to understand what’s happening in their business and make better decisions while those decisions can still have an impact.
The answers are already in your dealership.
Now you can just ask.
Lightspeed AI Command Center
Taylor Allis
Taylor Allis is Chief Product Officer at Lightspeed DMS, where he leads product strategy focused on simplifying dealership operations and applying AI to remove friction across the customer and employee journey. A seasoned SaaS and product leader, Taylor is known for translating complex technology into practical, customer-driven solutions. Before joining Lightspeed, he served as Chief Product Officer at Avetta, helping deliver scalable platforms that automated compliance, reduced operational overhead, and supported growth. Earlier in his career, he led Product and Marketing at Enablon and gained early innovation experience in the research labs at Sun Microsystems (later part of Oracle). Taylor is a named inventor on multiple patents and is passionate about building technology that helps businesses operate more efficiently and profitably. Outside of work, he enjoys coaching football and lacrosse—and when he’s off the clock, you’ll usually find him gravel biking, trail running, fly fishing, scuba diving, or hiking and climbing Colorado 14ers.
