Your Dealership Owns Its Data Now — What Custom Software Can Finally Do With It

Dealership DMS data is finally portable: $759.5 million in CDK and Reynolds antitrust settlements, Arizona-style data-access laws, and CDK’s 2026 Customer Data Platform mean dealer groups can now build their own cross-rooftop reporting, CRM sync, and inventory feeds — with custom integration work.

One clarification before anything else, because our own headline overstates it slightly. “Owns” is the marketing word. “Can finally get to it” is the accurate one. Who owns dealership data has been argued in contracts and courtrooms for a decade and is still argued today. What actually changed is access — and access is the only part a software build needs.

Why is dealership data suddenly portable?

Because three separate pressures landed on the same problem at roughly the same time: a decade of antitrust litigation that ended in $759.5 million of reported settlements, state data-access laws that survived their court challenges, and — most practically of all — the largest DMS vendor deciding to ship a data platform of its own.

Start with the litigation math. These figures are reported public record, per coverage from Bloomberg Law, Reuters, plaintiffs’ firm Milberg, and Auto Remarketing:

The $129.5 million dealer-class piece is the one most operators are still waiting on. The settlement administrator expects distribution to dealer class members before the end of 2026 — that is an expectation reported on the administrator’s own schedule, not a date to budget against. If a check is part of your plan, treat the timing as soft.

Now the part almost every vendor pitch gets wrong. None of those settlements created a data-portability right as a settlement term. They moved money and they closed cases. What they changed is the climate around access: after a decade of litigation over who may reach dealership data and at what price, the cost of telling a dealer group “no” went up and the cost of saying “yes” came down. Portability today is the combined effect of that pressure, of state laws that were already on the books, and of CDK’s own decision to productize data access. Anyone telling you a settlement handed you your data is selling you something.

What data rights does your dealership actually have?

Two sources, and most groups rank them backwards.

Your contract is the first source, and by far the strongest. Your DMS agreement, its integration addendum, and whatever certified-program terms you signed decide what you can pull, in what format, how often, and what it costs per rooftop. In a group that has grown by acquisition, those documents are frequently different at every store — the four-rooftop group that bought two stores in 2023 usually inherited two different sets of access terms and never re-read either. Before you scope a single line of integration work, someone in your organization has to be able to produce those contracts and point to the clause covering third-party access and export.

State law is the second source — and the leading example is much older than most people assume. Arizona passed HB2418 in 2019. The DMS vendors challenged it, and the Ninth Circuit denied their bid for an injunction in October 2021; the law stood. That is durable precedent and background, not breaking news. If someone has told you Arizona “just passed” a dealer data law, they are working from information that is five to seven years out of date. Its real value to you is what it established: a state can require DMS vendors to provide dealers and their authorized vendors with access, and that holding has been the reference point for every similar measure discussed since.

What we are not doing is telling you what your state requires. We build software; we do not practice law, and nothing here is legal advice. The settlements above are reported public record — QOS Software was not a party to any of them and has no stake in any of them. For what applies to your rooftops, ask your attorney or your state dealer association, then bring the answer to whoever is scoping the build.

What changed with CDK’s Customer Data Platform?

CDK announced its Customer Data Platform on January 22, 2026, launched it at the NADA Show, February 3–6, 2026, and began rollout in Q2 2026 with free migration for existing clients. Ford and Stellantis were named as pilots, per BusinessWire and CDK’s own newsroom.

It is a real product and it solves real problems: one unified customer profile across CDK’s own applications, fewer duplicate and stale records, and marketing activation that does not begin with someone exporting a spreadsheet at 6 a.m. If your group is all-CDK and your reporting needs live inside CDK’s footprint, this may genuinely be all you need. We will say that again in the closing section, because it is the most expensive mistake available in this space right now.

Here is what it does not cover:

• Rooftops that are not on CDK. A Reynolds store or a Dealertrack store inside your group sits outside the platform. Most multi-point groups are mixed, and mixed is exactly the case a vendor platform is not built to solve.
• Systems CDK does not sell. Group accounting consolidation, a legacy inventory feed to a marketplace, an OEM portal, a home-grown F&I or fixed-ops tool, the spreadsheet your CFO actually runs the group on.
• Terms you do not set. The schema, the retention windows, the roadmap, and the export rules belong to the vendor. That is not a criticism — it is the nature of every vendor platform — but it means the platform is not the place to keep data you want to survive a vendor change.
• History you want held independently. Ten years of deal and repair-order history in a system you own is a different asset from ten years of history in a system you subscribe to.

The platform’s predictive scoring and AI segmentation are the genuinely interesting part — and they are also the edge of our lane. QOS Software builds custom software. Standalone AI, agentic systems, and automation products belong to our sister company QOS Agentic. If the value you are chasing is the model rather than the plumbing, that is a conversation worth having, and it is not this one.

One more boundary, stated plainly: we are not a DMS and we do not replace one. CDK, Reynolds, Dealertrack and the rest run your stores. We build the software that sits beside them and moves data to where your group actually needs it.

What can a dealer group build with portable DMS data?

We have built software for dealership groups, and the pattern is consistent enough to predict: nobody’s first request is a data warehouse. The first request is always some version of “why can’t I see all seven stores on one page?” The warehouse is what you end up building because that question has no other honest answer.

• Cross-rooftop reporting. One view of gross, deal count, aged inventory, and fixed-ops performance across stores that run different DMS platforms — refreshed nightly instead of assembled by a controller every Monday. This is the payoff that justifies the rest of the list.
• A group-owned data warehouse. The only item here that survives a vendor change. Deal, inventory, customer, and repair-order history landed into a database your group owns, with your credentials and your backups, keeps its value whether you stay on your current DMS or leave it.
• CRM, inventory, and accounting sync. Inventory pushed to your marketplaces and your own sites from one source, customer records that do not have to be keyed twice, and deal data flowing into group accounting consolidation instead of being re-typed into a workbook.
• Service-drive triggers. Declined-service follow-up, warranty and lease-expiry alerts, equity notifications routed to the right advisor at the right store. These are simple rules that are impossible to run when the data is trapped in five separate systems and worth real money the moment it is not.

The general case for connecting your systems — why integration beats re-keying, and what a real integration project involves — is made on our API integrations page, and we are not going to rehearse it here. Whether you should build this or buy something close enough is its own decision, and that argument lives on our enterprise software development page. If you want the ROI-modeling method rather than the dealership specifics, our manufacturing software ROI post walks the same math in a different industry.

Is your dealership integration-ready?

Five checks. Run them before anyone quotes you anything — every one of them is cheaper to answer now than to discover in week three of a build.

If you score three or more rows in the “need work” column, the first phase of your project is not software — it is the contract audit and the data dictionary, and those are worth doing whether or not you ever build anything. The same sequencing logic applies to any multi-system integration; we walked through it step by step in our 3PL integration checklist, and the order of operations transfers directly.

What does dealership DMS data integration cost?

We are not going to invent a number for your group, because the honest answer is that four drivers set the price and three of them are decided before a developer opens an editor.

• Certified-program access fees. CDK and Reynolds each run their own certified-integration programs, and access to a DMS is generally a recurring per-store cost, not a one-time one. The pricing of that access is precisely what the antitrust litigation covered in the reporting cited above. Get the number in writing from the vendor, per rooftop, before you scope the build — it is an operating cost that outlives the project.
• Per-store scaling. Integration effort and access fees scale with rooftops, not with the group. A twelve-store group is closer to twelve small projects than to one large one, especially where DMS platforms are mixed. The savings come from reusing the mapping and the warehouse, not from a volume discount on access.
• Data-migration and normalization effort. This is the line item that surprises people. Reconciling what a field means at store A versus store E, cleaning duplicate customer records, and deciding how much history to bring forward routinely dominates the hours in phase one. Every group underestimates it, ours included, the first time.
• Ongoing maintenance. APIs version, schemas change, and a store switches DMS every few years. An integration is a system you own and maintain, not a thing you finish. Budget for it from day one rather than discovering it at the first breaking change.

For how custom development is priced generally — hourly, fixed-scope, retainer, and what actually moves a quote — we broke the whole model down in our custom software development cost guide for 2027.

Our take: when custom integration is the wrong move

Four situations where we would tell you not to build this — and we would rather tell you now than four weeks into discovery:

• You are a single rooftop. If one store’s DMS-native reports answer your questions, custom integration is an expensive way to reproduce something you already have. The value of this work is almost entirely in cross-rooftop visibility. One rooftop, no cross.
• You are all-CDK and the Customer Data Platform covers it. Free migration for existing clients is a real offer. Do not pay anyone to rebuild what your vendor now gives you. Pilot the platform first, find its actual edges, and only then scope what falls outside them.
• You are mid-DMS-migration. If a store is switching platforms in the next two quarters, every field mapping you build now gets rebuilt after the cutover. Wait. The one exception is extracting history off the outgoing system before you lose access to it — that work is timing-sensitive and worth doing early.
• Your data dictionary does not exist yet. If two stores cannot agree on what gross means, a warehouse will simply disagree faster and in more places. Fix the definitions first; that is a controller’s project, not a developer’s.

If none of those describe you — mixed DMS platforms, multiple rooftops, a real reporting question your current tools cannot answer — then the work is worth scoping, and the readiness matrix above is where to start. We have been building software under QOS since 2007, nearly two decades of shipped integration and data work across industries, and we package that capacity as blended senior-developer plus supervised AI capacity at founding-client launch pricing while our first clients onboard:

We also quote fixed-scope projects for groups that want one defined deliverable and one number — full tier details and current availability are on our pricing page. Either way, the first conversation is free and is mostly us asking about your five readiness rows: tell us what your group is trying to see, and we will tell you honestly whether it needs a build.

Frequently asked questions