DMS Cutover Checklist for Dealerships

Migration is a project. Cutover is a weekend. Most of the damage stores remember happened in about seventy two hours.

Migration is the months of work: choosing a vendor, mapping data, cleaning records, configuring the chart of accounts, training people. Cutover is the specific window when the old system stops being authoritative and the new one starts. It is usually a weekend, occasionally a month end, and it is where the stories come from.

The reason to separate them is that they fail differently. Migration failures are slow and visible. You find out in week six that your vendor cannot map a field, and you have time to react. Cutover failures are fast and public. A printer that will not print a buyer's order at 10am on Monday is not a project risk item, it is four salespeople standing around and a customer watching.

So the checklist below is deliberately narrow. It assumes your data mapping is done, your training happened and your contract is signed. If those are still open, start with switching dealer management systems and the DMS migration checklist instead, then come back here when your go live date is set.

One thing to state before anything else. LeadLocate is not a dealer management system. We do not do general ledger, accounts payable or receivable, payroll, bank reconciliation, deal posting to accounting, parts, repair orders, service scheduling, warranty claims or title and registration work. This page exists because our customers go through these weekends and ask us how to protect the sales side while it happens.

Everything on this list has to be settled before the week of, because the weekend has no room for a decision that needs three people and a phone call.

Name a single cutover owner. One person, with authority to stop the cutover. Not a committee. If your controller and your general manager both think they are in charge, you will discover it at the worst moment.

Set the go or no go criteria in writing. What conditions would cause you to abort and stay on the old system for another month. Decide this while everyone is calm. Stores that never define abort criteria never abort, even when they should.

Confirm your rollback position. Ask the vendor directly: if we stop on Sunday afternoon, what does it take to keep running on the old system on Monday. Get the answer in writing. Sometimes the honest answer is that there is no rollback after a certain point, and that is fine to know, but you have to know it.

Freeze non essential changes. No new users, no chart of accounts edits, no pay plan changes, no integration additions for two weeks either side. Every change added during a cutover becomes a suspect when something breaks.

Set the parts and service work in progress cutoff. Open repair orders crossing a conversion boundary are the single most common source of Monday chaos. Decide how they are handled and tell the shop before, not during. Related reading at parallel run conversions.

This week is about making the numbers true and the physical world ready.

Run your final data cleanup pass. Duplicate customers, dead vendor records, employees who left in 2022 and inactive accounts all migrate perfectly well if you let them, and then they are your problem forever. It is much cheaper to delete a record now than to explain it next year. There is more on this at data cleanup before migration.

Balance everything you intend to bring across, and print the proof. Schedules, inventory, floorplan, contracts in transit, accounts receivable aging, parts inventory value. Print or export the closing figures to a form that does not depend on the old system being available, because the day you need them is the day access has ended.

Do a test conversion and have the people who use the data check it, not the project team. Your parts manager will spot a wrong cost in ninety seconds. A project manager will not spot it at all.

Handle the physical layer, which is always underestimated. Forms and pre printed paper for the new system. Printer drivers and printer assignment for every station, including the one in the service drive nobody remembers. Scanners. Signature pads. Cash drawers. Label printers. Test each of them on the actual station with the actual paper. A form that prints half an inch off is a real problem when it is a legal document.

Finally, confirm what happens to historical records and for how long you retain access. Historical record migration and data ownership cover the questions to force answers on.

A workable shape for a Friday to Sunday conversion. Adjust the clock, keep the order.

1. Friday close. Final postings in the old system. Announce a hard stop time and enforce it. Everyone who thinks they can post one more thing after the stop is creating a reconciliation problem for your controller.
2. Friday evening. Final extraction. Confirm file sizes and record counts against expectation before anybody goes home. A short file at this stage is discovered cheaply and at 2am expensively.
3. Saturday morning. Load and verify. Check counts first, then spot check records by hand across every department. Do not accept a vendor's confirmation as verification.
4. Saturday afternoon. Balancing. Your controller compares the schedules to Friday's printouts. This is the true go or no go moment and it belongs to accounting, not IT.
5. Saturday evening. Permissions and users. Every person, every role, tested by logging in as them rather than by reading a settings screen.
6. Sunday morning. Devices and integrations. Print a real buyer's order, a real repair order and a real parts invoice on real paper. Confirm every third party that posts into or reads from the system, one at a time, by making it actually run.
7. Sunday afternoon. Go or no go against the criteria you wrote two weeks ago. Then communicate the decision to every employee, in writing, including what to do if something is wrong on Monday.

Build slack into every step. A conversion timed to finish at exactly the last available hour has no capacity for the one thing that always goes long. Conversion timelines covers realistic pacing.

The first Monday is not a normal Monday and pretending otherwise is how stores lose deals.

Staff the floor for slowness. Everything takes longer, so either book fewer appointments or add a body. Put your most experienced person where the pressure is instead of where the title says.

Have the vendor on site or on a dedicated line, agreed in advance. Not a support queue. A named person.

Set up a single visible issue list that anyone can add to, triaged twice a day by the cutover owner. Without it you get the same problem reported eleven times and one serious problem reported never.

Watch the money daily for the first week. Deals posting correctly, gross landing where it should, parts costing correctly, payroll data flowing. Do not wait for month end to find out. That is the controller's week and it is covered on the controller checklist.

Expect the emotional pattern, because it is universal. Day one is adrenaline. Day three is frustration. Day ten is quiet complaining that the old system was better. Day thirty is normal. If leadership panics on day three and starts talking about reversing, the store will never settle. Say all of this out loud before go live so it is a prediction rather than a crisis.

The items that are not on the vendor's project plan because they are not the vendor's problem.

Third party access. Every vendor with a connection to your old system needs to be repointed, and each one has its own lead time. Make the list four weeks out. It is always longer than anyone expects.

Bank and lender file formats. ACH files, floorplan feeds and lender funding uploads have specific formats that may change. Test one of each before the weekend, not after.

Scheduled reports and exports. The nightly file that feeds something in your marketing stack has a path that no longer exists. Nobody notices for a month, and then the report that stopped is the one somebody was making decisions from.

The stuff on people's desktops. Spreadsheets, saved queries, printed cheat sheets. They vanish quietly and take institutional knowledge with them.

Retention access to the old system. Negotiate read only access for a defined period before you sign, and know what your obligations are for keeping records. Discovering the retention requirement after access ended is a genuinely bad day.

Communication to customers. If your phone menu, your payment portal or your service reminders change, customers need telling. Silence generates inbound calls at exactly the moment your staff can least handle them.

This is where we have a genuine opinion, because it is the part our customers ask us about.

During a cutover, the accounting and fixed ops systems are unavoidably disrupted. What does not have to be disrupted is your ability to answer a lead, text a customer, book an appointment and desk a deal. If those live inside the system being converted, everything stops at once. If they live somewhere else, the store keeps selling while accounting sorts itself out.

LeadLocate runs independently. Neither an inventory feed nor access to the dealer management system you run is required to operate, which is deliberate rather than a limitation. Through a cutover weekend your lead inbox, distribution rules, SMS and MMS with RCS and SMS fallback, click to call with recording and transcription, voicemail drop, email, follow up processes, appointments and reminders all keep working. So does DealTracker desking with its fifty state tax matrix, so a deal can be structured and shown to a customer on a customer facing deal page even while your accounting system is mid conversion.

Two practical notes. Tell your team explicitly which system is authoritative for what during the window, because ambiguity is what produces double entry. And if you normally push desked deals into other systems, agree when that resumes rather than letting it happen halfway. DMS and CRM data synchronization covers the wiring in more detail.

Most aborted cutovers should have been aborted earlier and most disastrous ones should have been aborted at all. The difference is whether the criteria were written down before anybody was tired.

Useful criteria are binary and owned. Schedules balance to within a defined tolerance, and the controller says so. Every department has printed a real document on real paper. Every user has logged in as themselves. Every integration on the list has run once. Any one of those failing is a no go, not a discussion.

The pressure to proceed is always enormous, because the weekend was booked, people are exhausted and the vendor's team flies home Monday. Write your criteria as though a stranger will apply them, because by Sunday evening you will not be the same person who wrote them.

If you do abort, do it decisively and communicate it the same hour. A half abort, where some departments moved and some did not, is worse than either option.

And treat the vendor's confidence as an input rather than an answer. They have done this many times, which is genuinely valuable, but they will not be running your parts counter on Monday. Related questions to press during selection are on questions to ask during a DMS demo.

To be unambiguous after eight sections of somebody else's project. We do not sell a dealer management system, we do not convert one, and we do not do accounting, parts, service or title work. If a vendor tells you they can do all of that plus lead generation, read the contract carefully.

What we provide is the lead and communication layer that keeps producing while the rest of your stack is in pieces: exclusive local leads in a territory you choose, a full CRM with distribution rules, texting and calling with transcription, follow up automation, lead pages, personal salesperson websites, a free live chat widget, desking with customer facing deal pages, and secure credit applications with apply links through SecureWebX. Plus a phone validator and an email validator, which are unglamorous and save more wasted hours than anything else on the list.

Pricing is month to month with no long term contract: CRM Only from $199, exclusive local buyer lead programs from $799, Marketplace Acquisitions from $999 and the Buyers and Sellers Hybrid Plan from $1,599. Detail on the pricing page.

We cannot guarantee that your cutover goes smoothly, and neither can anyone else. What we can do is make sure the part of your business that generates money is not depending on the weekend going well. If you want to talk it through before your date, contact us.

How long should a DMS cutover weekend be?

Most stores plan Friday close to Sunday evening, with month end conversions being common. Build slack into every step. A plan that finishes at the last available hour has no capacity for the one task that always runs long.

Who should own the go or no go decision?

One named person with authority to stop, working from criteria written two weeks earlier while everyone was calm. The balancing check belongs to your controller, not to IT, because the numbers are the real test of a successful load.

What is the most common day one problem?

Printing. Forms, drivers, station assignments and pre printed paper. Test every device on the actual station with the actual document before the weekend ends, including the printer in the service drive that nobody thinks about.

Is LeadLocate a DMS, or does it convert one?

No to both. There is no general ledger, accounts payable or receivable, payroll, deal posting to accounting, parts, repair orders, service scheduling, warranty or title work. We provide the lead generation, CRM and communication layer alongside whatever system you run.

Can we keep selling during the conversion weekend?

Yes, if your lead handling, texting, calling, appointments and desking do not live inside the system being converted. Our platform runs independently and requires neither an inventory feed nor DMS access, so the sales side stays up while accounting converts.

What should we negotiate about the old system before signing?

Read only access for a defined retention period, a full export in a usable format including history, and written confirmation of what a rollback would require. Settle all three before signature rather than during your exit.

See a lead and communication stack that runs independently of your DMS, so a conversion does not stop your sales floor. Month to month, no long term contract.