If somebody handed me the keys to an empty dealership tomorrow, I wouldn't start by shopping for a CRM add-on, a giant TV dashboard, or some vendor promising "AI lead conversion."

I'd start with process.

Because that's the part everybody wants to skip.

People love talking about branding, traffic, inventory strategy, ad spend, and what software stack they're going to run. All of that matters. None of it saves you if the floor is sloppy, reps are winging it, internet leads sit for 18 minutes, and managers only find out what's broken after the store already missed the day.

I've spent enough mornings in a high-volume Kia store to know what falls apart first. It's not usually talent. It's not even traffic.

It's the lack of a system.

So if I opened a store from scratch, here's what I'd build first.

1. A Sales Process That Doesn't Depend on "Your Best Rep"

Before I hired the tenth salesperson, before I cared about the perfect word track, before I worried about whose name went on the wall, I'd build one clear road to the sale.

Not a motivational poster. An actual process.

What happens when an internet lead comes in? How fast does somebody respond? What counts as a real first contact? When does a manager touch the deal? When does the customer get T/O'd? What happens after an unsold showroom visit? Who owns the follow-up at 7pm if the rep is gone? What's the handoff to finance? What's the delivery checklist?

All of that gets decided up front.

Because if you don't define it, every rep creates their own version. One guy calls right away. One texts. One forgets. One pencils too early. One never asks for the appointment. And then leadership sits around later asking why the close rate is inconsistent.

You already know why.

I'd want every new rep to be able to walk in on day one and see the whole system. Internet lead process. Showroom process. Phone-up process. Unsold follow-up. Manager T/O standards. Delivery. Post-sale follow-up.

Nothing fancy. Just clear.

The best stores aren't magic. They're boring in the right places.

2. A Pay Plan That Rewards the Behavior I Actually Want

This is where a lot of stores lie to themselves.

They say they want great process, clean CRM, strong follow-up, solid gross, happy customers, and repeat business. Then they pay almost entirely on one thing and act surprised when the team ignores the rest.

Pay plans are culture in spreadsheet form.

So if I'm building from scratch, I want the pay plan to drive the right behavior from the jump:

I wouldn't build a comp plan that teaches reps to chase easy minis, skip notes, and dump a mess on the desk.

And I definitely wouldn't make it so complicated nobody can explain it without a calculator and a prayer.

Simple wins here.

A rep should know exactly how they get paid, what behaviors make them more money, and what habits cost them money. Managers should be able to coach directly from it. If somebody keeps missing T/O opportunities, that should show up in performance. If somebody sets appointments but never gets them to show, that should show up too.

This is one of those things people call "soft." It's not soft. It's the operating system.

3. A Daily Management Rhythm Before I Touch Any Automation

If the store opens at 8:30 and leadership doesn't know the truth until noon, you're already behind.

So day one, I'd build the rhythm of the day.

What numbers matter before the store opens? What gets reviewed in the morning huddle? What gets checked at midday? What gets reviewed before everybody leaves? Who sees it? What triggers action?

At my store, one of the biggest wins was replacing random spreadsheet digging with a morning report that lands before the day gets rolling. It hits my phone at 6:30am. I can see pace, inventory pressure, task load, and what needs attention before I even step onto the lot. That one change alone gave me back about 45 minutes a day. More important, it stopped me from walking into the store blind.

That's what I'd recreate first.

Not necessarily the same exact report. But the same discipline.

I want one scoreboard everybody can trust. I want yesterday's sold, month pace, lead response, pending deals, aged inventory, task pressure, and any fire that's already burning. I don't want five managers arguing over five different screenshots.

One source of truth.

And I wouldn't build it for vanity. I wouldn't care how pretty it looks on a dashboard if it doesn't change behavior. A good morning report should answer one question fast: where do I need to put my attention today?

If the answer is "nowhere," great. If the answer is "we have 26 unsolds with no manager T/O" or "internet response time slipped again," even better. Now we know where the leak is.

4. A Follow-Up Machine That Doesn't Let Deals Die Quietly

Most dealerships don't lose deals in dramatic fashion.

They lose them in silence.

The lead that sat too long. The unsold customer nobody called back. The trade lead that got one weak text and disappeared. The appointment that no-showed and never got worked again. The salesperson who thought they had it handled.

If I opened a store tomorrow, I'd build the BDC and follow-up system like my life depended on it. Because a big chunk of the store's profit does.

That means a few non-negotiables.

First, lead response time gets treated like a real metric, not a motivational slogan. If a fresh lead is sitting unanswered for more than a few minutes, somebody should know. Not at the end of the day. Right then.

Second, every unsold showroom gets an actual next step. Not "customer thinking about it." That's not a plan. That's a shrug.

Third, no-T/O deals get surfaced fast. One of the easiest ways to burn units is letting salespeople work an entire interaction without a manager ever getting involved. I've seen that one up close. It adds up fast.

Fourth, I'd run a short Save-a-Deal rhythm every day. Ten minutes. Tight. What's still alive, what stalled, what needs a manager call, what needs a different car, what needs a payment adjustment, what needs urgency. A lot of stores wait too long to go back and salvage the deal. By then the customer already bought somewhere else.

And fifth, the CRM can't become a graveyard.

I wrote before about staring at 847 overdue tasks in VinSolutions and realizing most of them were garbage. That's what happens when stores let the system pile up noise until nobody trusts it. If the CRM is full of junk, the important stuff gets buried with it.

I'd rather run a smaller, cleaner follow-up machine that actually gets used than a giant messy one everybody pretends to respect.

5. Inventory Discipline and Desk Discipline

This part isn't sexy either.

Still matters.

If I opened a store from scratch, I'd build rules around aging, pricing review, appraisal urgency, and desk logging immediately. Not three months later after everybody has their own bad habits.

Aged inventory can't be a monthly surprise.

Gross erosion can't be something we notice after the statement comes out.

Desk logs can't live in one manager's head.

I want every deal recorded the same way, every day. I want pricing review cadence. I want visibility into what units are getting old, what deals are getting buried in the desk, and where front-end gross is leaking.

And yes, I'd automate the stupid parts of that as soon as the process was stable.

The desk log is a perfect example. On paper it sounds minor. In real life it's 15 to 20 minutes of repetitive garbage every day, and more on Saturdays when you're fried and trying to leave. That's exactly the kind of work I want a machine doing.

But only after the structure is clean.

Automation on top of chaos just helps you create bad data faster.

6. Then I Add AI

Not first.

Then.

This is where the conversation gets weird because everybody wants AI to be the starting point. It's not. AI is the multiplier.

If the process is tight, AI becomes stupidly useful.

It can build the morning report before I wake up. It can clean CRM noise. It can triage email so I only see what matters. It can update the desk log. It can watch for metrics crossing a threshold and ping me instead of making me stare at dashboards all day. It can help me create content, prep training, and keep the dumb admin work from swallowing my morning.

That's the version of AI I care about.

Not the fake vendor stuff where a chatbot says "How can I assist you today?" on your website and everybody claps.

Real work.

At my store, the best automations save me a little over two hours a day. That's not theory. That's time I used to spend copying numbers, clicking through old tasks, digging through email, and typing into spreadsheets. Now that time goes back into the floor, the team, and the deals that actually matter.

If I were opening from scratch, I'd bake that in early. Not because AI is trendy. Because managers shouldn't be doing data-entry work in 2026.

What I Wouldn't Build Right Away

I wouldn't buy ten dashboards.

I wouldn't let every manager make up their own reporting format.

I wouldn't confuse traffic with process.

I wouldn't hand the floor to a couple "closers" and hope the rest of the team figures it out.

And I wouldn't spend six figures on software before I fixed the way the store actually works.

This is where people get tricked. They try to purchase discipline.

Can't do it.

A broken sales process inside expensive software is still a broken sales process.

If I Had to Boil It Down

If I opened a dealership tomorrow, my build order would look like this:

  1. define the sales process
  2. set the pay plan to reinforce it
  3. build the daily reporting rhythm
  4. tighten follow-up and Save-a-Deal recovery
  5. lock down desk and inventory discipline
  6. automate the repetitive junk on top of that foundation

That order matters.

Get it backwards and you spend the next year managing exceptions, cleaning up bad habits, and holding people accountable to standards that were never clear in the first place.

Do it right and the store gets easier to run before it gets bigger.

That's the whole point.