The Weekly Gross-Leakage Review That Assigns a Code and a Name to Every Soft Deal

Gross doesn't leak in one place, so it never gets fixed in one place

Every dealership has the same argument at the end of the month. Front gross is soft. Somebody says the market. Somebody says the used car manager overpaid on trades. Somebody says the desk gave away too much. Everybody's a little right, nobody owns it, and next month looks the same.

The problem isn't that people don't care. It's that gross leaks in four or five different departments, and the only place it shows up is one blended number on the DOC. By the time you see the average, the deals are cashed, the customers are gone, and the story is unrecoverable.

The fix is boring and it works: a 45-minute weekly review where you take every deal that came in under your gross threshold, assign one reason code, and assign one name. Not to punish. To create a pattern you can actually see by week three.

Pick the deals before the meeting, not during it

Do not walk into this meeting with a stack of deal jackets. Pull a list before you sit down.

The filter is simple: any retail delivery from last week where front gross landed below your threshold. Set the threshold where it captures roughly the bottom quarter of your deals — for most stores that lands somewhere between $500 and $1,200 front. If you're getting 40 deals a week and the list has 30 names on it, your threshold is wrong or you have a bigger problem than a meeting can solve.

Add three more buckets that don't show up as low gross but still cost you money:

• Units that went front-line ready more than 5 days after acquisition
• Appointments that showed and didn't get logged to a salesperson within 10 minutes
• Trades where the appraisal beat book by more than a set amount

Now you've got a working list of 8 to 15 items. That's a meeting.

The reason codes

Codes only work if there are few enough to remember and specific enough to argue about. Here's a starting set. Steal it, then cut anything you never use after a month.

D1 — Payment-driven discount. Customer had a number, we chased it. Common, legitimate sometimes, expensive when it's reflexive.

D2 — Competing quote. We matched or beat a written offer from another store.

D3 — Desk gave it early. Discount appeared before the customer asked or before a second pencil.

D4 — Manager exception. Somebody above the desk approved something outside policy.

T1 — Trade overallowance to close. We bought the trade high to make the payment work.

T2 — Appraisal miss. We bought it high because the appraisal was wrong — condition, mileage, market data stale.

R1 — Recon delay. Unit sat waiting on the shop, parts, or a sublet.

R2 — Recon overspend. Real cost came in materially over the estimate the appraisal was based on.

H1 — Handoff loss. BDC set it, customer showed, and the greeting or the assignment fell apart.

H2 — Data loss. The appointment existed but the salesperson didn't know what the BDC promised.

M1 — Market. Genuinely a unit or a segment nobody is paying for.

M1 is the escape hatch and everyone will want it. Set a rule: M1 requires a comp — a sold or a live listing that proves the point. No comp, no M1.

Assign a name, not a department

This is the part that most stores skip, and it's the part that makes the whole thing work.

Every coded deal gets one owner. Not "the desk." Not "recon." A person, by name, who is accountable for whether that pattern shows up again.

The owner isn't automatically the person who made the call. If a green salesperson gave up $800 because the desk never gave them a second pencil, the owner is the desk manager. If recon took 11 days because the parts order sat unapproved, the owner is whoever approves parts orders — not the tech.

Say it out loud in the room:

"Stock 4471, $400 front, code D3. Customer never asked. We came off $1,200 on the first pencil. Owner is me — I approved that pencil. Next week I want to see the first pencil on every desk log before it goes out on a unit under 30 days."

When the manager takes the first one, everyone else stops getting defensive.

What the meeting actually sounds like

Fifteen minutes of coding. Twenty minutes of pattern. Ten minutes of one commitment.

Coding goes fast because you're not relitigating deals. One line each:

You: Stock 3320, $250 front, trade came in $1,900 over book. Code? Used car manager: T1. Payment was $40 off and the customer had a lease turn-in date. You: Was there a second option — different term, different unit? Used car manager: We didn't build one. You: T1, owner is you. Log it.

Then you count. Out loud.

"Twelve coded deals. Five D3. Three T1. Two R1. One H1. One M1 with a comp."

Five D3 in one week isn't a salesperson problem. That's a desking policy problem, and it's the only thing worth fixing this week.

That's the third part: one commitment. Not five. Pick the code with the highest count, name the owner, and define what changes and how you'll check it.

"This week: no discount goes out on the first pencil without a manager initial. I'll pull ten desk logs Friday and count. If we're under two D3s next Tuesday, we keep it. If not, we tighten it further."

Where handoff losses hide

H1 and H2 will look small at first because they don't show up as a discount — they show up as a deal that never happened, and dead deals aren't on the gross report.

Pull them from the appointment log instead. Any confirmed appointment that showed and didn't turn into a written deal, cross-referenced with what the BDC actually promised. You'll find things like:

• BDC quoted a payment range the desk can't hit
• Customer showed asking for a specific unit that sold two days earlier
• Salesperson greeted them cold and re-qualified from zero, and the customer repeated everything they already told the BDC

That last one is the most expensive and the easiest to fix. A 30-second handoff — BDC walks the customer to the salesperson and says what they already know — recovers gross you'd never trace back to a code.

If you're already reviewing call and email recordings, this is where they earn their keep. The BDC's promise is on the recording. Compare it to what the desk could actually do, and H2 stops being a mystery.

Give it four weeks before you judge it

Week one you'll fight about codes. Week two people will show up with excuses pre-loaded. Week three you'll see a pattern nobody would have guessed — usually one manager, one lead source, or one vehicle segment producing a third of the leakage.

Week four is when the number moves, because you've been fixing one thing at a time instead of announcing that everyone needs to hold more gross.

Keep the log. Twelve weeks of coded deals is the most honest document in your store. When your GM asks why gross is where it is, you don't say "the market." You say: 41% D3, and here's what we did about it.