CDK Global Alternatives for Dealer Groups: 7 Options

This section is the reason this page exists, because the comparison pages a dealer finds when researching alternatives treat this decision as a features and star ratings exercise and do not mention any of it. Every statement of fact below traces to a named outlet, and each one is named inline rather than in a footnote. Where we could not read a source directly, we say so at that sentence instead of implying we did. Nothing here asserts or implies that CDK was found liable for anything, because it was not, and nothing here predicts how any pending matter resolves.

What happened

The attack. BleepingComputer, on June 22, 2024, reported that the BlackSuit ransomware gang was behind CDK Global's IT outage, dated the initial shutdown to June 19, 2024 with a second incident on June 20, and described BlackSuit as a May 2023 rebrand of the Royal ransomware operation, itself believed to be the direct successor of the Conti cybercrime syndicate. VERIFIED read directly off bleepingcomputer.com on August 29, 2026.

The scale. CBS News, in a story updated June 24, 2024, reported that the affected dealer management system is used by some 15,000 dealerships and powers sales for roughly half of the car dealerships in the United States, that CDK took its services down as a precaution, and that a second cyberattack that week compounded the problem. VERIFIED read directly off cbsnews.com on August 29, 2026.

The duration. TechTarget's explainer on the incident dates the first attack to June 18 and the second to June 19, and reports that after a phased restoration process all dealerships should have been running again by July 4, 2024. Different outlets place the first intrusion on June 18 or June 19, so we are giving you the range rather than a single tidy number: the outage began in the third week of June 2024 and was reported resolved by July 4, which is roughly two to three weeks depending on whose start date you take. VERIFIED read directly off techtarget.com on August 29, 2026.

The ransom. CyberScoop, on July 12, 2024, reported that researchers at blockchain intelligence firm TRM Labs identified approximately 387 bitcoin, worth roughly $25 million at the time, moving on June 21 to a wallet tied to the group, and that representatives for CDK and its parent firm refused to answer questions about whether CDK or a representative made a ransom payment. REPORTED read directly off cyberscoop.com on August 29, 2026. We state it as reporting rather than as fact because the company has not confirmed it, and because the researchers' own finding does not conclusively establish who sent the funds.

What it cost the industry

Anderson Economic Group, a consulting firm in East Lansing, Michigan, published a revised study estimating total direct losses to franchised auto dealers of about $1,020 million across the three calendar weeks that included June 19 and ran through early July 2024, including lost earnings on roughly 56,200 new vehicle sales that did not happen in the window, plus lost earnings on used vehicles, parts and service, additional staffing and information technology costs, and additional floor plan interest on inventory. Automotive News reported that revised figure under the headline that dealers lost $1.02 billion from the CDK cyberattack, and TechTarget independently reports the same firm's estimate as more than $1 billion collectively. REPORTED. Our direct fetch of both the Anderson Economic Group release and the Automotive News article was blocked, so we are relying on the study as reported by those outlets and on TechTarget's independent reference to it, which we did read directly.

Read that figure as what it is. It is a third party economic model of an industry wide loss over a defined window. It is not a CDK admitted number, it is not audited, and it does not tell you what your rooftop lost. Your own number is sitting in your June and July 2024 financial statements, and pulling it is the single most useful hour of preparation before any conversation about any of this.

What dealers were able to do about it

Lawsuits were filed quickly. The National Law Review, on July 18, 2024, reported complaints in the United States District Court for the Northern District of Illinois and the United States District Court for the Southern District of Florida, alleging among other things that CDK negligently failed to protect consumer information, with damages sought related to loss of commissions and sales. VERIFIED read directly off natlawreview.com on August 29, 2026.

A year later, they had mostly gone quiet. Automotive News reported, on June 19, 2025, under the headline that car dealers dropped or arbitrated their lawsuits over the 2024 CDK hack, that all of the federal civil lawsuits brought by retailers against the dealer management system provider appeared to have been dropped or to have gone to arbitration. REPORTED. We could not read the full article, which sits behind a subscriber paywall, so this is that outlet's published headline finding and nothing beyond it. If this matters to a decision you are making, have your counsel read the original.

One contract claim survived. Crain's Chicago Business, under the headline that CDK Global cyberattack lawsuits can proceed, and Automotive News, under the headline that a judge let breach of contract suits against CDK Global proceed, both reported that United States District Judge Jeremy Daniel allowed Travelers Excess and Surplus Lines Co., as subrogee of Mills Auto Group in Minnesota and Castle Automotive Group in Illinois, to pursue breach of contract claims arising from the 2024 attack, while dismissing the companion negligence and gross negligence claims under the economic loss doctrine, the rule barring tort recovery for purely economic losses arising from a failure to perform contractual obligations. REPORTED. Both outlets are paywalled and blocked our direct read, so this is their published reporting of the ruling and no more. Allowing a claim past a motion to dismiss is a procedural step and not a ruling on the merits. We make no prediction about how it resolves, and nothing here should be read as a finding that CDK breached any contract.

The part that is actually about you

Strip the headlines out and a pattern is left that has nothing to do with whether anyone behaved well. When a vendor agreement routes disputes to arbitration, a headline lawsuit rarely becomes a public verdict. Arbitration outcomes are typically confidential, which means that whatever other dealers recovered, you cannot look it up, cannot cite it, and cannot use it. Arbitration clauses are standard across software contracts generally and are not unique to this vendor or this category. That is not an accusation. It is the structure.

What follows from it is a practical instruction rather than a legal one, and it applies whether you stay or leave. Read your own agreement now, while nothing is on fire. Four things specifically: the dispute resolution clause and whether it sends you to arbitration; any business continuity, uptime or service level language and what remedy it actually provides if there is one; the limitation of liability clause, which in most software agreements caps recovery at some multiple of fees paid; and the data access, export and termination assistance terms. Those four paragraphs, not press coverage of what happened to somebody else, are what will govern your options on the worst day. Most dealer principals we talk to have not read them since signing. That is an hour of work with your own counsel and it is worth more than anything else on this page.

The thing we will not let this page imply

Commissioning a custom build is not a security strategy, and nobody should read this section as suggesting it is. A system you own can be breached. Moving off a large shared platform relocates the target and changes who is accountable, and that second part is the point: accountability moves to you and to whoever you build with. Name the obligations plainly, because they do not disappear when the vendor does. Backup cadence and, more importantly, tested recovery, meaning somebody actually restores from backup on a schedule and times it. Uptime monitoring and alerting that pages a human. Patch management on the operating system, the runtime and every dependency. An incident response plan that names who calls whom, in what order, with the phone numbers written down somewhere that does not require the system to be up. A custom build that skips any of those is less safe than a well run vendor platform, not more. And nothing about buying from us removes your own legal or regulatory exposure, which attaches to conduct and to the data you hold, not to who wrote the software.

Start with the uncomfortable finding, because it shapes everything after it. CDK publishes no price. We opened cdkglobal.com's own dealer management system page on August 29, 2026, it loaded normally with its full marketing content, and there is no pricing, no cost and no dollar figure anywhere on it. That is how we know rather than assume. Exactly one vendor in the table below publishes a real number on its own site. Six publish nothing at all.

Labels on this page are used in one fixed way, and we are copying the definition rather than paraphrasing it, because a paraphrase is how this word drifts. VERIFIED means read off the VENDOR's own page, or verified-by-absence at the vendor's own URL. A figure published by a third-party aggregator (Vendr, PriceLevel, SelectHub, ITQlick, G2, Capterra, checkthat.ai, any buyer-data or review platform) is REPORTED, however good that aggregator's data is. ASSUMPTION means it is a modelling input of ours rather than anyone's published figure. Where we found nothing, the cell says so instead of being filled in, and where our own attempt to read a source failed, the cell says that too rather than pretending the absence was observed.

Four numbers we deliberately did not print

This is worth a paragraph because the omissions are as much of the finding as the entries. Tekion. A specific monthly figure circulates in search summaries. We went to trace it and it attributed only to unnamed dealer level reporting with no source we could name, so it does not appear here. A number we cannot stand behind is worse than no number, because it looks the same as a real one on the page. Dealertrack. Cox Automotive's adjacent products do have reported figures in circulation, for vAuto and VinSolutions, but those are a merchandising tool and a CRM rather than the dealer management system, and printing them next to a DMS comparison would mislead. PBS Systems. Nobody has published a figure, vendor or third party. We checked the vendor's own site and it loaded, so the absence there is observed rather than assumed. Auto/Mate. The 65 to 75 percent claim on its own pricing page is a vendor marketing statement about its own product, not a verified comparison, and we present it as exactly that.

That leaves a category where one vendor out of eight publishes a checkable number and that number is zero. If price transparency is why you started reading, the honest finding is that switching does not deliver it. What switching can deliver is a different cost structure, and that is a different argument.

The Dominion DMS number, handled honestly

It is real and it is the vendor's own. Dominion DMS announced on January 27, 2023 that its VUE Core DMS platform would be offered to franchised automotive dealerships at $0 upfront and $0 per month, covering sales, service, parts, accounting and OEM communications through a web based interface, and stated an average savings of $6,500 per month per store against what dealers had been paying. That is VERIFIED because we read it on dominiondms.com, and the savings figure is the vendor's own claim about its own customers rather than an independent finding.

Now the catch, because $0 is not the same as free. The announcement does not price the optional third party integrations a real store adds on top of the free core, and monetising that layer rather than the core licence is the visible business model. So the honest reading is that the free core is genuine and vendor published, and it is not yet an apples to apples replacement for a loaded CDK bill until you have priced the add ons your store would actually need. Ask for that list in writing before anything else. It is still the most interesting number in this category by a distance, because it demonstrates that a core dealer management system can be delivered at a price of zero, which reframes what the licence half of your current bill is actually buying.

The five year model, and the caveat it carries

The model below takes a single rooftop, uses the REPORTED ITQlick base package figure of $1,995 per rooftop per month, adds the REPORTED $1,250 integration and training cost once at year zero, and escalates the licence four percent a year from the first renewal. That four percent is an ASSUMPTION when applied to CDK, and it is important to say why we chose it. CDK publishes no escalation term and no source gives a CDK specific annual increase rate. The one sourced escalation figure anywhere in this category is roughly four percent a year on Reynolds and Reynolds contracts, REPORTED by rfp.wiki. Rather than invent a higher number that would pull the crossover toward our own conclusion, we used the sourced figure from the nearest comparable vendor and labelled it. The five percent variant is printed below it so you can see how little the choice matters.

Against it sits a commissioned build at $45,000, which is the floor of our $45,000 to $180,000 range and not a quote for any specific dealer, with maintenance at fifteen percent of the build fee per year as a stated ASSUMPTION.

One year: $50,087.60 on CDK against $51,750.00 owned, and note the direction, because the build is still losing at that point. Two years: $75,981.10 against $58,500.00. Three years: $102,910.35 against $65,250.00. Five years: $160,043.43 against $78,750.00, a gap of $81,293.43.

Swap the four percent assumption for five percent and the same model gives $76,720.85 against $58,500.00 at year two and $164,087.79 against $78,750.00 at year five. The crossover year does not move. That is the useful thing about running both: the conclusion is not sensitive to the assumption we could not source.

Now run the loaded figure, and see why we did not lead with it

Put the REPORTED $30,000 per rooftop per month loaded figure into the same model and cumulative CDK spend is $361,250 before the end of year zero, against $45,000 for the build. The crossover is immediate, the five year gap is over $2.3 million, and the chart would be a vertical line next to a flat one.

We are showing you that case and not leading with it, for two reasons that are both about credibility. First, that figure is the weakest sourced number on this page: it comes from an aggregator, we could not load that aggregator's page ourselves on the day we built this, and it describes a fully loaded stack rather than a licence. Second, a crossover that fast reads as a sales pitch to any dealer principal who has actually negotiated a DMS contract, and they are right to distrust it. The conservative case is the one we will defend, and it is the one wired into the calculator's defaults. If your real invoice is closer to the loaded number, the calculator will tell you so as soon as you type it in, and the conclusion gets stronger without us having to reach for it.

The caveat that runs the other way. Both CDK figures exclude your third party data access fees, because those are charged per integration and nobody publishes a current schedule. If you are paying anything like the $200 per rooftop per month the DealerRefresh thread described in 2017, for each of several integrations, then every crossover on this page happens sooner for you than it does here.

Cumulative spend for a single rooftop, five years out, on the conservative case. CDK at the $1,995 per rooftop per month base package figure REPORTED by ITQlick, with the $1,250 integration and training cost from the same source paid at year zero, escalating four percent a year from the first renewal as a labelled ASSUMPTION benchmarked to the one sourced escalation figure in this category, roughly four percent a year on Reynolds and Reynolds contracts REPORTED by rfp.wiki. Against it, a commissioned build at $45,000, the floor of our $45,000 to $180,000 range rather than a quote for anyone, paid once at year zero, with maintenance at fifteen percent of the build price a year as a stated ASSUMPTION. Every figure in the chart comes from the table above, and every label comes with it. The loaded $30,000 per month case is not plotted because it would not fit this scale; it crosses over immediately, in year zero.

Look at the first two years before anything else, because that is the part most vendor charts hide by starting the axis somewhere flattering. The build line starts higher and stays higher through year one. At the twelve month mark the subscription has cost $50,087.60 and the build has cost $51,750.00, so staying on CDK is genuinely the cheaper decision for roughly the first thirteen months, which is where the two lines cross at about $52,000 of cumulative spend each. That is not a concession we are making reluctantly. It is the actual shape of the trade, and any dealer who needs the money to work inside twelve months should stop here.

The lines cross early in year two, at roughly $52,000 of cumulative spend on each path. After that the gap widens every single year, because one curve has a slope and the other is nearly flat. By the end of year three the difference is $37,660.35 in your favour. By year five it is $81,293.43, and nothing in the model bends the subscription line back down, because nothing in the real contract does either.

Change the build price and the crossover moves, which is the honest limit of the argument. At $45,000 it lands early in year two as drawn. At $95,000, the middle of our range, against this single rooftop it does not land until year six. At $180,000 against one rooftop it does not land inside ten years at all, which is as far as the calculator will run, and we would tell you that on the call rather than let you discover it. Add rooftops and it moves the other way fast, because the subscription is per rooftop and the build is not. That is what the calculator below is for.

Every default below is a figure from the table above, and every one of them is editable, because the defaults are third party estimates and your invoice is a fact. Nothing is submitted anywhere. There is no email gate, the tool makes no external request, and it stores no value. The arithmetic runs in your browser and stops there. If your inputs make the build lose, the tool says so rather than quietly hiding the result.

A word on the roster first. The alternatives lists a dealer finds for this product are mostly software directory pages that earn a fee when you request a demo, and it shows in the shortlists: consumer listing sites and independent lot tools sitting in the same table as enterprise franchised dealer systems, scored on star ratings, with pricing shown as Contact Vendor for almost every row. None of those pages mentions the outage, the contract structure, or what leaving actually involves. That does not make them worthless. It does mean the shortlist they hand you was not assembled for a franchised dealer holding a multi year agreement. The list below covers the systems CDK actually competes against for a franchise point, in the order a dealer would sensibly evaluate them.

1. Tekion Automotive Retail Cloud

What it is. The cloud native challenger, founded in 2018 and built after the cloud rather than migrated onto it, covering the full store on one platform. It is the system large groups shortlist when the complaint is that the incumbent feels architecturally old.

Price. VERIFIED that tekion.com/products/dms publishes no pricing; we loaded it on August 29, 2026 and there is no dollar figure on it. No third party figure appears here either, because the one in circulation traced only to unnamed dealer level reporting. Expect a multi month enterprise sales process before a quote.

Best for. Groups with the appetite for a full platform replacement and the internal capacity to run one, who want a single modern system rather than a core plus a decade of bolt-ons.

Where it falls short. Same pricing opacity as the incumbent, so it does not solve the transparency problem. It is also the destination in the one publicly documented data transfer fight in this category, described in the migration section below, which tells you the switch is not administratively simple.

Verdict. The strongest architectural upgrade on the list, and no more transparent about money than what you are leaving.

Is Tekion better than CDK?

Answered here as a section rather than spun into its own page, because it is a question inside this decision and not a separate decision. On architecture, most dealers who have run both describe Tekion as the more modern system, and that is the whole basis of its growth. On price, nobody can answer, because neither vendor publishes anything. On the switch itself, there is one documented data point and it is worth knowing before you start: Auto Remarketing reported on October 3, 2024 that the Superior Court of Gwinnett County in Georgia instructed CDK to provide the data of four Asbury Automotive Group dealerships to Tekion as part of a pilot program, and reported that before that order CDK had refused to transfer the data. Asbury's chief information officer was quoted saying the group was pleased it could proceed despite the roadblocks raised by CDK. REPORTED, read directly off autoremarketing.com on August 29, 2026. Tekion has separately alleged in its own public statements that CDK held dealer data hostage to block switching; that is an allegation made by a competitor which is also a party to litigation against CDK, and we present it as an allegation and nothing more. The practical takeaway for a dealer is neither vendor's marketing. It is that the data exit is the part of this switch to get in writing first.

2. Dealertrack DMS (Cox Automotive)

What it is. Cox Automotive's dealer management system, sold as a cloud platform with open integrations and positioned heavily on data fee policy, and in practice bought alongside the rest of the Cox stack: vAuto for inventory, VinSolutions for CRM, Xtime for service scheduling.

Price. VERIFIED that Dealertrack publishes no price for the system. We loaded us.dealertrack.com/content/dealertrack/en/dealer-management-solutions.html on August 29, 2026, with its feature list and its FAQ answers, and there is no price on it. The only dollar amounts on that page sit inside a data fee savings estimator, which is a marketing widget rather than a rate card. No third party figure for the DMS itself could be sourced either. Pricing is negotiated dealership by dealership and is frequently bundled with other Cox products, which makes the effective rate harder to isolate rather than easier.

Best for. Dealers already deep in the Cox ecosystem, and dealers for whom third party integration policy is the central complaint, since Dealertrack markets openness on exactly that axis.

Where it falls short. Bundling is a double edged benefit. A blended discount across four products is genuinely cheaper and also genuinely harder to unpick at renewal, and it deepens single vendor dependence rather than reducing it, which is the opposite of what a dealer shopping after June 2024 usually says they want.

Verdict. The most natural switch for a Cox house, and a lateral move on the concentration risk that sends most people looking.

3. Reynolds and Reynolds

What it is. The other legacy incumbent, selling ERA-IGNITE and POWER under its Retail Management System umbrella. Between them, Reynolds and CDK have held the franchised dealer market for decades.

Price. No published pricing. REPORTED at $1,000 to $5,000 per store per month depending on modules, typically on three to five year contracts with roughly four percent annual increases, per an rfp.wiki dealer management system comparison carried in this site's own pricing ledger. We are not claiming verified absence at the vendor's own site: reyrey.com returned only page titles to our reader on August 29, 2026 rather than readable body content, so we did not observe the absence ourselves and we will not label an unread page.

Best for. Dealers who want the other established incumbent, with comparable original equipment manufacturer certification depth and a comparably mature support organisation.

Where it falls short. This is a lateral move on every axis that this page is about. Same pricing opacity, same long contract terms, same concentration. Both companies were also named as defendants in the dealer antitrust class action that CDK settled for $100 million in August 2024, over an alleged conspiracy to charge unlawful prices for dealer management system services and data integration services, per the Milberg firm's published account of the settlement, which states expressly that the settlement is not an admission of wrongdoing or liability by any party. REPORTED, and stated as an allegation resolved by settlement without admission, not as a finding against anyone.

Verdict. A real option that changes your vendor and not your situation.

4. DealerSocket

What it is. A CRM led platform now inside Solera, with dealer management adjacent modules including inventory, desking and equity mining. The name gets used for several different products, which is worth knowing before you compare quotes.

Price. No published pricing; dealersocket.com/pricing returned a Page Not Found error when we checked it on August 29, 2026. REPORTED third party cost guides put the base at $750 to $4,000 per store per month, with onboarding reported at $3,000 to $15,000 and integrations far above that. Treat that range with suspicion in the useful sense: it is wide because it spans different products under one brand, so confirm which module any quote covers.

Best for. Dealers whose real pain is the CRM and sales process rather than accounting and factory reporting.

Where it falls short. It is not a full franchised dealer management system replacement in the way CDK Drive, Reynolds or Tekion are, and buying it as though it were is a well documented way to end up running two systems and paying for both.

Verdict. Strong in its own lane. Do not shortlist it as a like for like DMS swap without confirming exactly which modules are in the quote.

5. Auto/Mate

What it is. A dealer management system that built its reputation on being the anti incumbent option for independent minded stores, with a support led culture. Also now part of Solera.

Price. VERIFIED that automate.com/dms/pricing/ publishes no dollar figure; we loaded it on August 29, 2026 and it directs buyers to a sales conversation. The page does carry one comparative claim, which we quote as the vendor's own marketing statement rather than as an independent finding: dealers often find their Auto/Mate monthly support bills to be 65 to 75 percent less than what they are paying with Reynolds and Reynolds or CDK. No independent source confirms that, and no dollar figure exists anywhere to test it against.

Best for. Small and mid size franchised dealers who feel they are paying enterprise pricing for a store that is not enterprise scale.

Where it falls short. A vendor's own percentage claim about its competitors' bills is a marketing number, and the absence of any dollar figure means you cannot check it before entering a sales process. Consolidation under a larger owner is also a legitimate diligence question for a dealer buying on the strength of a support culture.

Verdict. Worth a quote if your bill feels disproportionate to your store size. Bring your invoice and make them show the comparison on your actual numbers.

6. PBS Systems

What it is. A North American dealer management system running the v10 APEX platform, covering accounting, sales, service, parts and CRM, with a substantial installed base in Canada and a growing one in the United States.

Price. VERIFIED that pbssystems.com publishes no pricing; we loaded it on August 29, 2026 with its full navigation and it directs buyers to request a demo. No aggregator, forum or third party source publishes a dollar figure either, which we checked rather than assumed. That combination is worth naming plainly: any number you are quoted cannot be tested against a public benchmark, because there is no public benchmark.

Best for. Single point stores and small groups, particularly with Canadian operations, who want a genuinely different vendor rather than the other half of the duopoly.

Where it falls short. Smaller third party integration ecosystem than the incumbents, which matters more than dealers expect, because the tool you like may not connect. And the total pricing opacity means the negotiation is entirely on their information.

Verdict. A real alternative outside the duopoly. Get the integration list for the tools you already run before anything else.

7. Dominion DMS

What it is. The VUE platform, and the only vendor in this roundup that publishes an actual number on its own website.

Price. VERIFIED, read directly off dominiondms.com on August 29, 2026: the VUE Core DMS platform is offered to franchised automotive dealerships at $0 upfront and $0 per month, covering sales, service, parts, accounting and OEM communications through a web based interface, announced January 27, 2023. The vendor states an average savings of $6,500 per month per store against what dealers had been paying, which is its own claim about its own customers rather than an independent finding.

Best for. Dealers whose licence line is the problem, and dealers who want a concrete number to hold every other quote against. Even if you never buy it, a published $0 changes what you can ask everyone else.

Where it falls short. The announcement does not price the optional third party integrations a real store adds on top of the free core, and monetising that layer is the visible model. So $0 is the honest headline for the core and not the honest headline for your total bill. Get the add on schedule in writing before you compare anything.

Verdict. The most interesting data point in the category, and the strongest evidence that the licence half of a loaded incumbent bill is not the part doing the work.

8. A commissioned build you own

What it is. Not a dealer management system, and we will keep saying that because the distinction is where most bad projects start. It is a custom system built for the workflows no dealer management system covers the way your store actually runs them, sitting alongside whichever DMS you keep, owned by you at handoff. In practice that means things like service scheduling and follow up that behaves the way your best advisor behaves, an internet lead response layer that answers in seconds at eleven at night, equity and service to sales mining against your own book on your own rules, or the deal jacket and compliance document chase that your finance office rebuilds by hand every week.

Price. A fixed fee of $45,000 to $180,000, set after a free 45-minute diagnosis and after integration depth is named, paid in two installments at build start and at handoff. You own the code, prompts, models and pipeline at handoff and run it in your own cloud tenant. Maintenance is modelled at fifteen percent a year on this page as a stated ASSUMPTION, not a standing contract term.

Best for. Dealers and groups whose DMS basically works and who can name one or two workflows that leak money every month. Multi rooftop groups get the most from it, because the fee does not multiply by rooftop and the subscription does.

Where it falls short. It is a bigger single cheque, it needs a tighter scope than buying software does, and it does not make your DMS go away. If what you want is to stop paying CDK entirely, this is not that. It is also not a security strategy, as the outage section above says at length.

Verdict. The option nobody in the table above will show you, and the only one where the bill stops going up.

Each of these is either arithmetic you can check on this page or a structural fact about the two models. No invented percentages appear in any of them. Where an argument does not honestly apply to your store, the section right after this one says so.

One. The math, restated. A subscription never ends. On a single rooftop, modelled at the $1,995 base package figure REPORTED rather than published, escalating at the four percent ASSUMPTION benchmarked to the one sourced figure in the category, against a build carrying maintenance at the fifteen percent ASSUMPTION, cumulative CDK spend passes cumulative build spend early in year two: $75,981.10 against $58,500.00 at the twenty four month mark. By year five it is $160,043.43 against $78,750.00, a difference of $81,293.43. Year six on the subscription starts from zero progress. Year six on the build starts from an asset you already own.

Two. Per rooftop pricing taxes growth. Every point you add raises the licence on day one whether or not it performs in year one. At the reported base figure that is $23,940 a year per rooftop, added automatically, with nobody making a decision. A commissioned system has no marginal licence cost per rooftop at all, which takes a software line item out of every acquisition model you build. For a group that intends to buy stores, this is the argument that compounds hardest.

Three. Asset versus expense. Five years of subscription payments leave you with nothing on the balance sheet and no leverage in the next negotiation. A build is a transferable asset. If you sell the store or the group, the buyer acquires a working system rather than inheriting an assignment negotiation with a vendor. Nobody buys a dealership because of its DMS, but buyers do discount for operational dependencies they cannot control, and this is one.

Four. Built around your store, not the other way round. A dealer management system is built for the average of fifteen thousand dealerships. Your pay plans, your service drive process, your used car reconditioning workflow and your factory's specific reporting quirks are not average, which is why every store runs a layer of spreadsheets and habits beside the system. A commissioned build starts from that layer instead of asking you to abandon it, and you pay for none of the modules your store never opens.

Five. AI at the core, not sold by the seat. Across this category, artificial intelligence arrives as a module quoted on top of the base contract, and not one vendor in the table above publishes what that module costs, which we checked rather than assumed. In a commissioned build the AI is not a line item at all, because it is the system. That difference compounds with headcount on one side and not on the other.

Six. Unlimited seats. Add a salesperson in March, three porters in June and a seasonal service greeter in November, and an owned system costs the same as it did in February. Per seat and per rooftop software makes headcount a software decision, which is exactly backwards for a business whose staffing moves with the season and the market.

Seven. Data ownership, and no exit ransom. This is the argument with the strongest documentary support in this vertical, and it is not ours. Auto Remarketing reported on October 3, 2024 that a Georgia court instructed CDK to provide four Asbury Automotive Group dealerships' data to Tekion for a pilot, after CDK had refused to transfer it. REPORTED. Whatever the merits, note the mechanism: a publicly traded dealer group needed a court to move its own data. With a commissioned build the data sits in your cloud tenant under your credentials, and there is no counterparty to ask.

Eight. Vendor risk you stop carrying. Consolidation is the constant in this category, and two of the seven alternatives on this page are now inside the same parent company. Products get acquired, merged, repriced and sunset on somebody else's schedule, and your migration timetable becomes their roadmap decision. An owned system is not on anybody's roadmap but yours.

Nine. Change speed. A change request to your own system is a conversation and a sprint. A feature request to a platform serving fifteen thousand dealerships is a queue position behind fourteen thousand nine hundred and ninety nine other stores, and the honest answer from any vendor is that your priority is a function of your size. Dealers who have sat through a two year wait for a workflow change already know which of those two things they are buying.

Ten. One fixed fee, and proof you can check. $45,000 to $180,000, scoped, one time, set after the diagnosis call. On proof, we will only name what is real and checkable. Jim Glaser Law is a named client and the principal takes reference calls. We built and run the LELF platform. Our AI systems have handled more than 6,000 calls in production. We have delivered more than 40 commissions. We have not built a dealer management system, we have no auto retail case study to show you, and we would rather you read that here than discover it on the call.

Six situations where every argument in the previous section fails, and where we would tell you to stay put on a call.

Single point stores where the crossover is genuinely marginal. Read the chart again. At one rooftop on the conservative figure, the build is still losing at the twelve month mark and only pulls ahead early in year two. If your store is at the smaller end and your invoice is at the low end, the arithmetic is close enough that the operational disruption is not worth it, and the answer is to renegotiate rather than build.

Groups with deep institutional knowledge and in house CDK administrators. A controller and a service director who have run this system for a decade are fast in it in ways that do not show up on a comparison table. Retraining that team has a real cost. Where consolidation across many rooftops is genuinely working, the burden of proof sits on the change rather than on staying.

Dealers whose factory requirements are the binding constraint. Original equipment manufacturer certification and reporting requirements are not optional and not uniform across brands. If your franchise agreements point at a shortlist of certified systems, that shortlist is the decision, and no amount of cost modelling overrides it. Confirm certification for every brand you hold before evaluating anything.

Stores already mid transition, mid acquisition or mid construction. A DMS migration is a real operational risk stacked on top of whatever else is happening. Doing it during an acquisition integration or a facility rebuild is a way to have two problems. That is a reason to time the move, not to rule it out.

Dealers whose actual pain is one feature gap. If the honest answer to why you are looking is one missing integration or one clunky screen, a full platform migration is disproportionate on any measure. Raise it with your account team in writing first and see what comes back before anyone builds a business case.

Dealers with nobody to own a system internally. A commissioned build needs one named person who cares about it, even at a light touch, and in a dealership that person is usually the general manager or the controller and is already fully occupied. Without them you get an orphaned system that decays quietly, which is a worse outcome than renting. If you cannot name that person today, rent.

1. Have you read your own agreement's dispute resolution, liability, uptime and data access clauses in the last twelve months? If no, stop here and do that first, with counsel, before evaluating a single vendor. It is the highest value hour available to you and it costs nothing but the hour. Everything below assumes you know what your contract says. If yes, continue.

2. Do your franchise agreements restrict you to a shortlist of certified systems? If yes, that shortlist is your decision and cost modelling does not override it. Stop, and take the total cost figure from this page into your renewal instead. If no, or if the shortlist has more than one name on it, continue.

3. Do you have your actual CDK invoice, including every third party data access line, in front of you? If no, stop and go find it. Everything after this depends on your real number, and the reported figures on this page are a poor substitute for that document rather than a replacement for it. If yes, continue.

4. Run your numbers in the calculator above. Is your five year CDK total below about $110,000? That $110,000 line is our own rule of thumb rather than anyone's published figure, ASSUMPTION: the cheapest build we would scope is $45,000, which is $78,750 over five years with maintenance at our stated assumption, and we set the cutoff meaningfully above that because a margin that thin does not justify the disruption. If your total is under it, stop; renegotiating at renewal is the better use of your energy. If no, continue.

5. Is the thing costing you money actually a dealer management system problem? If yes, meaning core accounting, factory reporting, parts inventory or deal structure, then a different DMS may genuinely help, and Tekion, Dealertrack, Reynolds or PBS Systems is where to look, with Dominion DMS as the number to hold them all against. Stop here and go price them. If no, continue.

6. Can you name the workflow in one sentence, with a rough dollar or hour figure attached, and is the budget for a $45,000 to $180,000 fixed fee real this quarter? If no to the first half, stop and spend two weeks measuring before anyone spends money; every failed build we have seen started with an unnamed constraint. If no to the second half, park it and revisit at renewal with the total cost figure from this page in the file. If yes to both, the diagnosis call is the next step, and a meaningful share of them end with us telling a dealer to stay where they are.

Almost nobody writing about switching answers this with specifics, which is strange, because it is the question that decides whether a dealer ever acts on any of the rest.

The data exit is the project, and there is a documented case that proves it. Auto Remarketing reported on October 3, 2024 that the Superior Court of Gwinnett County in Georgia instructed CDK to provide the data of four Asbury Automotive Group dealerships to Tekion as part of a pilot program, and reported that prior to the entry of that order CDK had refused to transfer the dealerships' data to Tekion, which competes with CDK in providing dealer management systems. Asbury's vice president and chief information officer was quoted saying the group was pleased it could proceed with the pilot despite the roadblocks raised by CDK. REPORTED, read directly on August 29, 2026. We are describing what a named outlet reported about a court order and a party's public statement, not making a finding about anyone's conduct.

Sit with the scale of that for a second, because it is the whole argument. Asbury is a large publicly traded dealer group with counsel on retainer, moving four stores as a pilot, and the transfer still went through a court. Whatever your own agreement says, plan the data exit as the first work stream rather than the last, and get the export format, the fields included, the historical depth and the delivery timeline in writing before you sign anything with anyone.

The fee structure around data access has its own long history. Two separate antitrust matters resolved without any admission of wrongdoing, and both are about data access fees rather than about the outage. In August 2024 CDK agreed to pay $100 million to a class of United States dealerships that bought a CDK or Reynolds dealer management system between September 2013 and August 2024, over an alleged conspiracy by CDK and Reynolds and Reynolds to charge unlawful prices in the markets for dealer management system services and data integration services, per the Milberg firm's published account, which states that the settlement is not an admission of wrongdoing or liability by any party. In January 2025 CDK agreed to pay $630 million to settle claims by AutoLoop and 242 other software vendors who alleged CDK cut off access to auto dealer systems and drove up the prices vendors pay to access data for their applications, filed in federal court in Madison, Wisconsin, with the company denying any wrongdoing in agreeing to settle, per Claims Journal. REPORTED, both read directly on August 29, 2026. These are a different legal track from the ransomware matter and should not be blended with it: the $1 billion loss estimate and these settlement figures are not the same money and are not about the same events.

What actually has to move. Not one database. Customer and prospect records with their full contact and consent history. Deal jackets and finance and insurance documentation, which carry retention obligations. Repair order history, which your service retention marketing depends on and which is the thing dealers most often discover they lost. Parts inventory and pricing files. Accounting history and the general ledger, at whatever depth your accountant and your factory require. Payroll and pay plan configuration. Factory communication interfaces, which have to be recertified rather than migrated. And the report set your general manager reads every morning, which is the one everybody forgets and the one that gets noticed on day one.

Timelines, stated honestly. We did not find a source for a CDK specific migration duration that we would put a number behind, so we are not printing one. Anyone quoting you a tidy figure for this should be asked where it came from. What we will say, and label as our own view rather than a sourced figure, is that the phases are predictable even when the duration is not: discovery and data audit, cleanup, configuration, multiple test loads, parallel operation, cutover, and a stabilisation period before you decommission anything. Ask any vendor bidding for the work to put their own timeline in writing with the phase breakdown attached, then hold them to it.

What drives the number is your data, not your destination. A store with a decade of inconsistent charge codes, duplicate customer records and three people maintaining the parts file differently will take the long end of any range regardless of which platform it picks. You can measure this in a week by asking your controller how many manual adjustments it takes to close a month, and your service manager how much of the repair order history they would actually trust.

What we do and do not do here. We have not run a CDK decommissioning and we are not going to imply otherwise. What we build is the workflow layer that sits alongside whichever system you land on, which is a different job. If your project needs a migration partner, that is a specialist engagement, and the platform you are moving to will usually name two or three.

The option most dealers do not consider. You do not have to leave to fix the problem. In a large share of the cases we see, the platform is fine, the licence is defensible, and the money is leaking in a workflow sitting beside it: internet lead response after hours, service scheduling and follow up, equity mining, or the document chase in the finance office. That path keeps the DMS, keeps the institutional knowledge, keeps the factory certifications, and builds only the missing piece. No migration, no parallel run, no data extraction fight.

For the pricing evidence behind this page and every other category we track, the vertical software pricing index carries the dated ledger, including its auto dealers and repair cluster. If your operation runs trucks as well as showrooms, McLeod Software alternatives runs this same sourced cost model on the transportation side, and ServiceTitan alternatives does it for field service, which is the closest structural analogue to a fixed operations department. For the underlying decision rather than the vendor comparison, renting AI versus owning it lays out what renting really costs against a build over the same horizons, and off the shelf versus a custom commission is the five test framework we run on every diagnosis call, including the tests that send people back to buying. The full comparison library holds the rest, how the fixed fee is set explains where a number inside $45,000 to $180,000 comes from, and the process page walks through what the weeks after a signature look like.