Who owns dealership data? Legally, in most agreements, the dealer does. Functionally, the vendor does. If you cannot export your full customer data set in a usable format without a fight, a fee, or a six-week delay, you do not own your data. You generate it. Someone else controls it.
That distinction is the most expensive fine print in automotive retail.
The extraction
Here is the arrangement most dealers have signed up for without ever naming it.
Every deal you log, every vehicle you service, every campaign you send produces data. That data flows into vendor systems. The vendors aggregate it across thousands of rooftops, refine it into benchmarks, scores, and algorithms, and sell the refined product back to the industry. Sometimes back to you. Sometimes to your competitor across town.
You generate the raw material. They own the refinement.
That is not a partnership. That is extraction.
No single contract created this. It accumulated, one integration fee and one data clause at a time, until the average store was paying $150K to $400K a year to vendors who know more about its customers than it does.
The one-question test
You do not need a lawyer to find out where you stand. You need one question and a close eye on the reaction.
Ask your DMS or CRM vendor: "Can I export my full data set?"
Then watch what happens. The answer reveals who actually owns your customer relationships.
If the answer is yes, today, in a standard format, at no charge, you own your data. That answer is rare.
More often you get a version of: yes, but. But there's an export fee. But it takes 30 to 60 days. But it comes as flat files with no relationships intact. But certified integrations only. But the enriched fields stay with us.
Every "but" is a fence post. Count them and you have mapped your cage.
Rented intelligence vs. owned intelligence
The ownership question matters more now than it did five years ago, for one reason: AI runs on data, and whoever holds the data holds the intelligence.
Renting intelligence means your insights live inside a vendor's model, trained on aggregate industry data, and they stop the day you stop paying. Twenty years of customer history and on cancellation day you walk away with flat files and start over.
Owned intelligence compounds every day you operate. Your data history becomes an asset no competitor can replicate, because they cannot buy your last fifteen years of ROs, deals, and customer behavior at any price.
This is the asymmetry to internalize: vendors churn, tools get replaced, platforms come and go. The data layer is the only part of your technology stack that should outlive every vendor relationship you have.
What bad ownership actually costs
The costs are not theoretical. They show up in the P&L wearing disguises.
Duplicate records. Paid twice for the same customer. Wasted ad spend. Eroded trust when the same household gets three conflicting offers. Stores lose 3 to 5 percent of gross revenue annually to bad data.
Unreachable contacts. In the average dealership, 42 percent of CRM contacts are unreachable. Wrong numbers, dead emails, moved households. You are marketing to ghosts and paying full rate for the privilege.
The RO gap. The average repair order writes up at $900 and collects $494. Closing that gap requires clean history and connected systems. Fragmented vendor data keeps it open.
Bad attribution. Mispriced units. Deals that make no sense at month close. You cannot improve what you cannot accurately measure, and you cannot measure across systems you do not control.
How to take it back
Taking back the data layer is a sequence, not an event.
First, audit. Pull your duplication rate from the DMS. Most stores find 20 to 40 percent. That number is your baseline and your business case.
Second, test your rights. Run the export question past every vendor holding customer data. Write the answers down. You are building the map of what you actually control.
Third, define "customer" in writing. One definition, agreed by leadership. Most stores cannot answer it the same way twice. Every downstream system inherits this definition, so settle it before you build anything.
Fourth, build or buy a data layer you own. One environment where DMS, CRM, inventory, marketing, phone, and website data get cleaned, deduplicated, validated, and stored under your control. Vendors connect to it. They do not own it. Data flows both directions, and the system of record is yours.
From there, every AI decision gets easier. The vendor question becomes simple: what data are you running on, and do I own it? Vendors who can't answer clearly are renting you their intelligence.
The stakes
The dealers who win the next five years own their data layer. They don't rent it from vendors. Their AI runs on operational-grade data only they possess. Their intelligence compounds while their competitors' resets with every vendor switch.
The window to make this move is open now, while most of the industry still treats data as exhaust instead of equity. It will not stay open. Whoever knows their customer deepest wins the deal, and that depth is built on ownership.
FAQ
Does my dealership legally own its DMS data? Most agreements say yes on paper. The functional test is whether you can export your complete data set in usable form without fees, delays, or missing fields. Many dealers cannot.
What should I ask my DMS or CRM vendor about data? "Can I export my full data set?" Then, for AI vendors: "What data are you running on, and do I own it?" The reactions tell you more than the contracts.
How much does bad data cost a dealership? Roughly 3 to 5 percent of gross revenue annually, through duplicate records, unreachable contacts, missed service opportunities, and bad attribution.
What is the difference between rented and owned intelligence? Rented intelligence lives in vendor models and stops when you stop paying. Owned intelligence is built on your data layer and compounds every day you operate.
What is the first step to owning my data layer? Audit your DMS duplication rate. Most stores find 20 to 40 percent duplicates. That baseline becomes the business case for everything that follows.
Todd Smith is the Founder and CEO of QoreAI and the author of The Intelligent Dealership: How AI and Data Transform Automotive Retail.
Free resource: The Dealer Data Addendum is an ungated set of eight contract clauses (data ownership, export rights, deletion, schema-change notice, audit rights) you can hand to your attorney and attach to any vendor agreement.