By Michael Quigley, President & Chief Strategy Officer, Co-Founder, Impel
Walk any showroom and you’ll find a business that has turned inventory into a science. Days supply, turn rate, acquisition cost, aging. Every unit gets measured, priced, and moved with intent.
Now look for that same rigor applied to the asset with the highest ceiling of them all. The one that never touches the lot.
That asset is data. And how a dealership manages it is fast becoming the difference between a business that compounds and one that quietly leaks value every day it opens the doors.
Buyers are already telling you what they want
The latest Cox Automotive Car Buyer Journey Study makes the signal hard to miss. Among buyers who engaged AI tools during the shopping process, 84% reported high satisfaction, among the highest of any buyer group in the survey. 83% of consumers say AI will reshape car buying. 63% of dealers agree that investing in AI now is critical to long-term success.
That last number is worth sitting with. Nearly two-thirds of dealers already understand the imperative. But understanding and executing are different things. And the gap between them almost always comes down to the same root cause: the data underneath the AI isn’t ready.
The inventory analogy that actually holds
Inventory management didn’t become a science by accident. Dealers learned that the right car, priced right and stocked at the right moment, was the whole margin. Data sits in exactly that position today.
A car loses value every day it sits unsold. Customer data loses value every day it sits siloed and unused. Communication history. Service records. Purchase patterns. Channel preferences. Intent signals. Lifecycle stage. Every one of those pieces is an asset. And every one of them behaves like a unit on the lot, appreciating when actively managed, depreciating when left alone.
The question was never whether a dealership owns valuable data. Every dealership does. The question is whether that data is working or quietly going stale in the dark.
The layer most dealers are skipping
Here’s what the research makes clear that most AI conversations in automotive skip past entirely.
The 2026 Stanford HAI AI Index Report found that 74% of organizations now cite data inaccuracy as their top AI risk, up 14 percentage points in a single year. That puts data quality ahead of cybersecurity, regulatory compliance, and privacy as the number-one concern among AI leaders globally. The risk isn’t AI failing. It’s AI running on bad data and executing confidently in the wrong direction.
Gartner’s May 2026 analysis goes one layer deeper. Organizations that prioritize semantics, meaning context, in their AI-ready data will increase agentic AI accuracy by up to 80% and reduce costs by up to 60%. The analyst put it directly: context with semantic coherence will become a cost-control and trust strategy, not a nice-to-have.
That finding reframes the whole conversation. The question isn’t whether a dealership has AI. It’s whether the AI has context.
Context is what turns a generic follow-up into a relevant one. It’s the difference between an AI that knows a customer’s name and an AI that knows their vehicle, their service history, their lease end date, their last declined repair, and their preferred communication channel. Generic AI can produce a response. Contextually rich AI produces the right response, the one that feels like it came from someone who actually knows the customer.
That’s the concierge experience. And it only exists when the data underneath it is unified, current, and semantically coherent.
The fragmentation problem nobody voted for
The data keeps compounding whether anyone manages it or not. Volume was never the problem. Fragmentation is.
In the typical store, customer data is scattered across the DMS, the CRM, the marketing platform, the service scheduler, the chat tool, and the phone system, with none of them speaking the same language. The AI sitting on top of that fragmentation isn’t getting smarter with every interaction. It’s getting confused.
Picture running vehicles across five lots with no shared system to track any of them. No dealer would tolerate that for a week. Yet that’s exactly how dealership data tends to end up.
The moat that can’t be copied
Physical inventory is replicable. Any dealer can stock the same cars, run the same promotions, match the same price by Friday. Data is the exception.
A dealership with five years of rich, unified, semantically coherent customer data holds something no competitor can buy, borrow, or rebuild overnight. Put well-configured agents to work on that data and the advantage compounds year over year, because as I’ve written about in Centaurs, Chess, and the State of Vertical AI in 2026, every interaction the system handles makes it smarter. Every month it isn’t running is a month it isn’t learning.
And as I argued in The Down Escalator Problem, the cost of delay isn’t linear. The dealership that fixes its data layer now builds a moat that widens with every interaction. The one that waits inherits a gap that grows harder to close.
The takeaway for dealers
Start treating data the way you treat your most valuable franchise. Manage it with intent and it appreciates. Neglect it and it depreciates, quietly and expensively, every single day.
The path forward starts with three questions about your data:
- Is it unified? A dealership’s data should flow from one customer record that every agent, every department, and every touchpoint draws from and contributes back to. Five systems running five versions of the same customer isn’t a data strategy. It’s a liability.
- Is it current? Data that isn’t actively maintained depreciates. Service history from two years ago without recent engagement signals isn’t a foundation. It needs constant refreshing to stay useful.
- Is it contextually rich? Name and contact info is table stakes. Vehicle history, service behavior, lease end date, declined repairs, and communication preferences — that’s the semantic layer that separates a generic AI response from a concierge one.
Unified. Current. Contextually rich. That’s the foundation. Build agents on top of it and the moat widens with every interaction. Without it, even the best AI can only do so much, because the quality of the output will always reflect the quality of the input.
63% of dealers already agree that AI investment is critical. The ones who win won’t just be the ones who invested. They’ll be the ones who built the data foundation that made the investment mean something.
________
Michael Quigley unpacks shifts like this one every issue. Subscribe to Drive the Future for the thinking before it becomes the consensus.
Frequently asked questions
What is a dealership data strategy?
A dealership data strategy is the deliberate plan for how a dealership collects, unifies, governs, and activates its customer data. It treats data as a strategic asset, much like inventory, rather than a byproduct of daily operations. A strong strategy pulls fragmented sources into one customer view that staff and AI agents can both act on.
Why does data quality matter for automotive AI?
Data quality determines how well AI agents perform. Agents working on thin, fragmented data deliver generic, low-trust interactions. Agents working on rich, unified data deliver personalized, high-converting ones. IDC notes that organizations without high-quality, AI-ready data risk roughly a 15% productivity loss scaling AI by 2027.
How is customer data like dealership inventory?
Customer data behaves like inventory because it appreciates or depreciates based on management. A car loses value the longer it sits unsold. Customer data loses value the longer it sits siloed and unused. Both reward the same discipline: the right asset, activated at the right moment.
What is an AI-infused data architecture?
An AI-infused data architecture unifies a company’s data into a clean, connected foundation that AI systems can reliably use. IDC predicts 40% of organizations will invest in these architectures by 2027 to protect decision quality and competitiveness. For dealers, it means one data layer feeding every agent and every department.
How does Impel unify dealership data?
Impel’s AI Operating System runs Sales AI, Service AI, Voice AI, Marketing AI, and VinVision AI on a single, shared data layer. Every agent draws from the same customer view and contributes back to it. That unified foundation lets dealers activate their data consistently across sales, service, and marketing.