The Silent Leak
A straight-talk guide for franchise dealers who know their follow-up is broken and haven't found anyone they trust to fix it.
Who I wrote this for
You run a single point store. Maybe franchised, maybe an established pre owned operation that's been in the same family for two generations. You do somewhere between 60 and 180 units a month. A hundred or more internet leads come in every month, and there are years of contacts sitting in your CRM that nobody has touched since the week they came in.
You've worked the desk. You sign the checks. And you've personally watched a lead die in your own system with nothing you could do about it in the moment.
That's who this is for.
Who I did not write it for
If you run a large dealer group with a centralized BDC and a director whose whole job is response time, this guide is not written for you. Your version of this problem is a different animal. Leads die in your stores too, but the thing nobody can tell you is which rooftop is letting them die, or what the spread between your best store and your worst one is actually costing you. That's a real conversation and I'm glad to have it. It just isn't this one.
If you're buy here pay here, this doesn't fit your model.
And if you're looking for the system that puts twenty extra units on your board every month forever, put this down right now. That's not what this is, and anybody selling you that is going to disappoint you. I'd rather lose you on page one than on month three.
Still here? Good. That by itself puts you ahead of most of the market, because the hard part of this particular problem is that it never announces itself and almost nobody goes looking for it.
The 9:15 lead
Tuesday night. 9:15. Somebody sits on their couch with a phone and fills out a form on your website about a used unit on your lot. They give you their real name and their real cell number, because they actually want the car.
Nobody at your store sees it that night. Fair enough, you're closed.
Wednesday morning, one of two things happens. Either somebody picks it up and calls, or it goes into the pile with the rest of the internet leads and gets a task reminder that turns into another task reminder. By Wednesday afternoon that customer has already heard back from the store across town and is sitting in a demo.
You paid for that lead. You lost that deal.
Here's the part that actually costs you money: you will never know it happened. In your CRM that customer looks exactly like the other two hundred leads that didn't convert this month, which is to say they look like nothing at all, because there is no alert and no flag and no line anywhere in your reporting that says this one was ready and we let them go.
The aged pile in your CRM is the same wound, just older. Those are the same people, the ones who raised their hand and got silence and went and bought somewhere else, and some number of them are back in the market right now without any idea that your store still has their name sitting in a database.
That's the leak. It's quiet, it's daily, and it doesn't show up on any report you currently run.
So I went and checked
I spent 25 years in this business, most of it in an F&I office. I've watched how leads get worked in the real world, which is not the same as how the process manual says they get worked.
So when I started LeadCentrix I did what anybody does. I wrote up what I do, printed it, and started walking into stores with a smile and a business card. I got niced out the door in about forty seconds, over and over.
That's fine! I sold Kirby vacuums for seventeen years before I ever sold a car, so I know that drill and I don't take it personally. But it told me something I needed to hear: nobody was going to take my word for any of this, especially not from a stranger. I needed records. And truthfully, I wanted to find out for myself whether the thing I'd believed for twenty five years was actually true, or whether it was just something we all say to each other in the finance office on a slow Saturday.
How I ran it
Over several weeks I shopped a handful of dealerships, the same way one of your customers would. Here's the method, so you can judge it for yourself:
- Monday evenings, after hours. That's when a real buyer is home and browsing, and it's the window where stores are most exposed.
- I picked a specific vehicle off the store's own website. Every time, I chose a used unit that had just aged out of its factory comprehensive coverage.
- I varied the channel. Some weeks I called. Some weeks I submitted the web form. When I called, I used the number that came up on a plain Google search, not the tracked number buried on the site.
- When I submitted online, I used the most obvious option on the page - whatever the store's own site put in front of me first.
- Then I tracked every response for the next five days. Every call, text, email, voicemail. Timestamps on all of it.
The vehicle choice was on purpose. A used unit just past its comprehensive coverage is the most natural service contract conversation on the lot, and a customer asking about one has already started thinking about what ownership is going to cost them. Twenty five years in F&I will do that to you. I wasn't shopping a random car. I was shopping the kind of lead you and your finance office both want walking through the door.
The Google number matters too, and I'll come back to that one.
I want to be straight with you about the size of this. It was a handful of stores in one market over a few weeks. It's a field check, not a national study. I'm telling you what happened, and you can decide what it's worth.
And none of it is complicated. Every bit of what I just described costs nothing except a week of paying attention, which means you can run the identical test on your own store starting tonight. Frankly I'd rather you did that than take my word for any of what comes next. There's a checklist at the back to make it easy.
What came back
Less than you'd think. A lot less.
Across the stores I shopped, most of them produced no human contact at all inside the first 48 hours. Not a call, not a text from a person, nothing but whatever automated reply the website fired off on its own.
One store never made a single human attempt in five days. That store has a staffed BDC. I want to sit on that one for a second, because a priced unit and a real cell number and a real name went into a store that employs people whose entire reason for existing is to work exactly that lead, and five days later not one of them had picked up a phone!
Before you file that under somebody else's problem, it's worth knowing your own number. Pull your last twenty after hours leads and count the human attempts on each one. There's a full checklist at the back of this guide.
Another store did almost everything right at the start. Fast reply. Asked permission before texting me, which is more than most do and I'll give them full credit for it. Then a call and a voicemail the next morning.
And then it stopped. That was the whole follow up. One call, one voicemail, done.
That store is running a modern CRM with a built in messaging assistant, and as best I can tell the technology did exactly what it was installed to do. The follow through still ended after one attempt.
That's the finding I'd underline if I were you. A store with a modern CRM and a messaging assistant still stopped after one call. Whatever you've got installed is not the same thing as persistence, and it's not the same thing as knowing which lead turned into which deal.
The rest went silent from the first minute. No automated reply, no call, no email, nothing at all - a form submission on their own website went into a system somewhere and never came back out of it.
I want to be careful with what I'm saying here. This isn't a story about a handful of bad dealerships. Not one of those stores did anything unusual, and every one of them is run by people who work hard and think their leads are being handled. That's the point. This is what normal looks like in a business where almost nobody has ever measured it.
The answer I got in every room
On the Fridays after each week of shopping, I walked into those stores with the screenshots and the phone records and asked to talk to a manager. Not to pitch. To show them what happened to a customer on their own website.
Every one of those conversations happened off to the side, out of earshot of anybody else. That part was on purpose. No manager should have to hear that in front of their people.
Every single one of them had an explanation. Every one!
- "You called the wrong number."
- "You used the chat box instead of the lead form."
- "You didn't inquire correctly."
That last one came from a sales manager standing in their own showroom, telling a live prospect that the store's own website chat is not the correct way to contact that store.
Read that again, because it's the whole book in one sentence.
Not one manager looked at the records and said "that shouldn't have happened, let me find out why." Every one of them pointed outward. And honestly, I get it. A stranger walks into your store on a Friday afternoon holding printouts that say your people let a ready buyer go, and there is no version of that conversation where the first thing out of your mouth is thank you.
About that wrong number
I want to take the "you called the wrong number" answer seriously, because there's something buried in it that should bother you more than it bothered the person who said it.
I called the number that came up on a Google search. That's the number a real customer finds. That's the number on the map listing, the one in the search result, the one your own community has had in their phones for years.
So when a store tells me I called the wrong number, what they're telling me is that calls arriving on their most publicly visible phone line land outside whatever they measure.
If that's true at your store, then some unknown share of your inbound is invisible to you. Not badly handled. Invisible. You can't manage a number you never see, and you can't fire a vendor for a lead you never knew arrived.
That's not a follow up problem anymore. That's an attribution problem, and it's quietly making every other number you look at wrong.
One more thing about those managers
Not one of them was lying to me. That's what I keep coming back to. They believed their stores were handling leads, because as far as they could see, their stores were handling leads. Nobody walks the floor looking for the customer who isn't there.
And remember where those conversations happened. Off to the side, nobody listening. There was no audience to perform for and no face to save, which means the answer I got was the honest one as that manager understood their own store. That understanding was just missing a piece nobody had ever put in front of them.
That's the reason this is a guide instead of a sales call. Nobody's standing in front of you right now. There's nothing to defend and nobody watching you read it, so you can look at your own store without an audience, which is about the only condition under which any of us ever look at anything honestly.
If you're this far in, you're already doing the thing those managers never got the chance to do.
"We don't hammer or pester people"
Of all the answers I got, this is the one worth the most of your time. A manager at one of those stores said it to me plainly, and they weren't embarrassed about it. It's their standard. One call, a voicemail, and then leave the customer alone.
They meant it as a virtue. I don't think they're a fool for it.
Let me make their argument as well as they would make it. Customers are sick of being hounded. Everybody's phone is full of junk texts. Their name is on the sign in a town where their family has been for forty years, and they are not going to let their store become the one that blows people up. They would rather lose a deal than earn that reputation.
That's a real position, held by a competent operator. If you've said something close to it, you're in good company.
Here's the problem with it.
The customer asked you to call
The person on the other end of that inquiry raised their hand. They typed their own phone number into your website at nine o'clock at night. Whatever else is true, they were not an interruption. They started it.
One call and a voicemail into silence does not read as restraint on the receiving end. It reads as nobody wanted the business. That customer isn't at home thinking what a considerate dealership that was. They think they got ignored, and they buy from whoever called back.
And notice what this belief does to your reporting. When a store's standard is one attempt, a lead that dies after one attempt looks like a success. Nothing failed. Everybody followed the process. That's exactly why a store can have a BDC on the payroll, a modern CRM in the rack, and a customer who never heard from a human being - with not one person in the building thinking anything went wrong.
The standard is the leak.
Persistence and pestering are two different things
They're right that volume for its own sake is obnoxious. Seven texts in nine days that all say the same thing, sent by something that doesn't know it's already been answered - that's what they're protecting their store from, and they should.
But that's not what a second touch has to be. The difference comes down to two things:
Every touch carries a reason. New information, a real question, something that wasn't in the last one. A message that just says "checking in!" is the pestering they're talking about. A message that answers the question the customer actually asked is a different animal.
The sequence has an end, and the customer gets told. A plain "I haven't heard back, so I'm going to close this out unless you tell me otherwise" is about the least annoying message you can send, and it gives the person something to answer. Not one of the stores I shopped ever sent it. They just stopped, which leaves the customer with no signal at all.
So we don't argue with the manager. We agree with them and then we build the version that actually respects the customer: a reason for every touch, a defined ending, and a hard stop the second somebody says no.
What it's actually costing you
I'm going to give you numbers, and I'm going to tell you exactly where each one comes from. If I can't source it, it's not in here.
That's not me being noble. There's a close rate number that gets passed around this business that I used for a while myself. When I went looking for the study behind it, I couldn't find one. So I threw it out and started over.
The two rates that do the work
Urban Science looked at dealership leads and sales across the summer of 2025 and measured what closed inside 30 days. Internet leads came in at 6%. Phone leads came in at 14%.1
Now, those are not the same people and I'm not going to pretend they are. A customer who picks up the phone and calls your store has already decided they want to talk to somebody, and some of that gap is nothing more than that.
Here's what's underneath it anyway. A form and a phone call can come from the same person, about the same car, on the same Tuesday night. Which one your store ends up holding depends on whether anybody there turned the first one into the second one.
That's the job. Not answering the form. Getting that person into an actual conversation while they still care about the car, because a conversation is where a deal starts and a form is where one waits.
Everything after this is about closing the distance between those two numbers.
Speed is the whole ballgame
Researchers at Harvard Business Review studied thousands of online leads and found that response time inside the first hour is a different world from response time after it. Firms that reached a lead within an hour were far more likely to have a real qualifying conversation than firms that waited even a couple of hours.2
Now go back to the 9:15 Tuesday lead. If nobody touches that one until somebody rolls in Wednesday morning at nine, you are not a little bit late, you are hours outside the window where the research says the conversation was still there to be had.
The number most dealers leave out
When you total up what a dead lead costs, you probably think in front end gross. That's not the whole bill.
Average F&I income per retail unit ran $1,727 in Q1 2026, per the Presidio-NCM Average Dealership Performance Benchmark Report.3
That money rides along with the deal, which means that when the deal dies in silence the F&I gross dies right along with it, and it never shows up as a loss anywhere in your statement because you cannot book a loss on a car you never sold.
Run your own version. Take your monthly internet lead count against a 6% close rate, then ask yourself honestly what share of those leads got a real human attempt inside an hour. The gap between those two numbers is what I'm calling the leak - and every unit in it carries a front end gross and an F&I gross with it.
The part that gets skipped
Before anyone touches your aged database, somebody has to answer a question that's easy to wave off as paperwork: is this contact safe to message at all?
I'm not going to explain the law to you. I'm not a lawyer, and that conversation belongs to yours. What I can do is show you the public record and then tell you exactly what we built because of it.
| Business | Year | What was claimed | Reported amount |
|---|---|---|---|
| Grieco Ford of Fort Lauderdale | 2018 | Unsolicited robocalls and texts4 | $4.8 million |
| Moss Bros. Auto Group (CA) | 2022 | Prerecorded telemarketing messages without express written consent5 | $2.5 million |
| Toyota of North Miami | 2025-26 | Texts sent after customers opted out6 | $889,525 |
| Classic Chevrolet + 8 OK dealerships | 2020 | Telemarketing texts without consent, including numbers on the DNC registry7 | $850,000 |
| JK Buick GMC (IL) | 2024 | Texts sent after customers asked them to stop8 | $130,000 |
Each of these settled. None of the dealers admitted wrongdoing, and a settlement is not a finding of liability. The figures are as publicly reported.
Look at the range: $130,000 up to $4.8 million. A single point Buick store in Illinois sits on that list next to a large Florida Ford operation. Size is not the filter.
Three of those five involve the same claim. Messages that kept coming AFTER somebody said stop.
That tells you where the real failure lives. It isn't the first text. It's a follow up system with no reliable memory of who opted out, so the next campaign or the next tool sends message eleven to a person who quit at message three.
Two more details from those records are worth your time. In one case the class covered messages sent by the group's employees, agents AND third party contractors. In another, the class definition specifically included texts sent through the dealerships' texting vendor.
Put those together and the lesson is plain. When a vendor sends on your behalf, those messages are treated as yours.
What we do about it
Here it comes... the pitch! HA!
I'd rather say that out loud than pretend this next part isn't one. You've read six chapters of me telling you what's broken, so hopefully I've earned the right to lay out what I built. You decide for yourself whether it's any good. Then we'll get back to your store.
Every message in our system clears five checks before it can leave:
- A documented consent record that names your store, verified before we send.
- A DNC and reassigned number scrub, in case that number changed hands since the record was made.
- Quiet hours in the customer's local time, held to the most restrictive window. Never Sundays or holidays.
- A hard frequency cap, on any subject.
- No opt out on record. Any "stop," in any wording, is permanent. "Don't text me" counts. "Wrong number" counts.
Fail one, the message dies. There's no override and no "probably fine" pile.
Our suppression list never expires, and it survives every campaign, every export and every tool change. If a question ever comes up later, it resolves as a records question. We hold the records.
This costs us messages, and I'm not going to pretend it never costs a deal. A contact we set aside might have been a live buyer. But losing one possible deal is a small price to pay for avoiding time in front of a judge.
What an honest fix looks like
Two pieces, in order. The first one proves the gap was real. The second one closes it.
First: recover what's already in the building
We take your aged leads, screen them the way I described, and re-engage the ones that pass. Compliant, on your store's behalf, with a reason behind every touch and a defined end to every sequence.
When somebody responds and wants to talk, we warm them up to the point of scheduling and expecting a callback from your team. Your people get them in, car them down and close the deal. That's their job, and they're good at it.
You pay a flat $500 when a deal actually closes. Nothing before that. No setup fee, no monthly minimum, no retainer, no data fees. We cover the carrier registration cost too.
Second: stop it from re-forming
Recovery gets back what you already lost. It doesn't stop tomorrow's 9:15 lead from dying the same way.
That's the ongoing side: rapid response on new inbound around the clock, after hours coverage so the nine o'clock caller gets handled instead of dumped to voicemail, and rolling re-engagement so nothing ages into the pile again. A flat monthly operating fee per rooftop, all in, with no per deal charges stacked on top.
What we need from you
Almost nothing, and that's deliberate.
A spreadsheet. That's it. No software to install, nobody to hire, no login to your CRM or your DMS, no access to your financials or your deal data. Your team's workflow doesn't change. They just start getting warmer conversations handed to them.
Your salespeople make more money when the leads they're handed are already interested. Nobody on your floor loses anything in this arrangement, which is worth saying out loud before somebody on your team assumes otherwise.
What this won't do
You've been sold to before. So here's the part most of those pitches left out.
Reactivation is not a machine that prints extra units every month forever. Your database is a finite pool. We're mining an asset you already own, not manufacturing new demand. The first run is the strongest one. After that it tapers, because we've already worked the people who were there.
We don't close anything. We warm a prospect to the point where they're expecting a call from your store. Everything after that is your team, your process, your gross. If your closers can't close, we can't fix that and I won't pretend to.
We can't work leads that don't have clean consent. Some of your database is going to come back set aside, and depending on how your forms were built over the years, it could be a meaningful chunk. I'd rather tell you that up front than surprise you with it.
It's not instant. Registration with the carriers takes time before a single message goes out. We handle it and we eat the cost, but I'm not going to tell you we're live on Tuesday.
So what's it actually good for?
Two things. Real deals recovered out of leads your store already wrote off, and something you can't get any other way: a hard, documented number for what your follow up gap has been costing you. That second one is the reason the first one matters. The recovery run proves the size of the hole. Then you decide what to do about it.
Five things you can check this week
You don't need me for any of this. Run it yourself. If everything comes back clean, throw this guide in the trash with my compliments!
- 1Shop your own store. Tonight, after close, from a phone nobody there recognizes. Pick a used unit off your own website and use whatever contact option your site shows a customer first. Then wait, but don't respond. Write down every single thing that happens and when. Do not tell anybody you're doing it.
- 2Count the attempts, not the tasks. Pull your last twenty internet leads that came in after 6 PM. For each one, count actual human attempts - a call placed, a text a person typed, a voicemail left. Automated emails don't count. My guess is your average is closer to one than three.
- 3Call your Google number. Search your store the way a customer would, call the number that comes up, and then go look for that call in your reporting. If you can't find it, you've got leads arriving where you can't see them.
- 4Measure the pile. Run a count of every lead in your CRM older than 90 days with no activity since. Not to work them. Just look at the number. It's the cheapest thing on this list to check and the hardest one to un-see.
- 5Pull one consent record. Pick any contact in that aged pile and try to produce the documentation of what they agreed to and when. If you can find it in under five minutes, you're in better shape than most. If you can't, that's worth knowing before anybody texts that list - us or anyone else.
Those five checks will tell you more about your store than any vendor presentation will, including mine.
Or send me fifty and I'll do Step 5 for you
Step 5 is the hardest one to do yourself, and it's the one that decides whether any of this is even possible at your store.
So here's the free first step, and there's nothing attached to it.
Export 50 of your aged leads to a spreadsheet. Contact detail, and the consent record behind it - what they agreed to and when. Nothing else. No financials, no notes, no deal history, and no login to anything.
Email it to 50-leads@leadcentrix.com. That address goes straight to the review, so you don't have to get past anybody to use it.
We review all fifty the same way we would before a live campaign, and you get back a record by record accounting: which ones are workable, which need a second look, which we'd set aside, and the reason for each one.
You keep that report whether you ever work with us or not.
It's an audit, not a pitch. And I'll tell you right now, part of what comes back is going to be me telling you what we can't touch and why. That's the part that protects you.
Start the 50-Lead Proof
No setup fee. No retainer. No CRM access. Your next deal is closer than you think.
Send the spreadsheet to 50-leads@leadcentrix.com Or call (850) 495-7561 · www.LeadCentrix.comP.S. If you only do one thing from this whole guide, do Step 1. Shop your own store tonight. Whatever comes back is the truth about your follow up, and it costs you nothing but the ten minutes.
P.P.S. If you're sitting there thinking you've heard a version of this before from somebody who couldn't back it up, you're right and I don't blame you. That's why the first step is free, requires no access to your systems, and produces a document you keep either way. You find out what I'm worth before you give me anything.
About John Dooley
I spent 25 years in the car business, most of it as an F&I manager, before I built LeadCentrix. I've watched how leads actually get worked, which is why none of this guide is theory.
LeadCentrix is an operational infrastructure company. We're not an agency, we don't sell leads, and we're not an outsourced BDC. We recover revenue out of databases dealers already paid for, and we get paid when a deal closes.
Pensacola, FL. Serving dealerships nationwide.
Sources
- Urban Science, "Close Rate vs. Defection Rate: What Both Metrics Reveal About Dealership Performance." Internet leads averaged a 30 day close rate of 6 percent, phone leads 14 percent, based on May through July 2025 lead and sales data. urbanscience.com/resources/close-rate-vs-defection-rate-what-both-metrics-reveal-about-dealership-performance/
- James Oldroyd, Kristina McElheran and David Elkington, "The Short Life of Online Sales Leads," Harvard Business Review, March 2011. hbr.org/2011/03/the-short-life-of-online-sales-leads
- Presidio-NCM Average Dealership Performance Benchmark Report, Q1 2026, as reported by WardsAuto: average F&I income per retail unit of $1,727, up 7.1% year over year. wardsauto.com/news/fixed-ops-fi-carry-the-load-for-dealerships-in-q1/818466/
- Automotive News, "Florida dealership settles robocall suit for $4.8 million" (Dec. 2018). autonews.com/article/20181210/RETAIL07/181219965/
- Johnson v. Moss Bros. Auto Group, Inc. (C.D. Cal.), final approval June 24, 2022. Orrick InfoBytes. infobytes.orrick.com/2022-06-29/district-court-approves-25-million-settlement-over-prerecorded-telemarketing-messages/
- Karpiel v. FRL Automotive LLC (d/b/a Toyota of North Miami), Fla. 11th Jud. Cir., No. 2025-020201-CA-01. Preliminary approval Nov. 24, 2025; final approval hearing held Mar. 25, 2026; claims period closed Apr. 9, 2026. ClassAction.org. classaction.org/news/889525-toyota-of-north-miami-settlement-ends-class-action-lawsuit-over-alleged-marketing-texts
- Jennifer King v. Classic Chevrolet Inc., et al., No. 4:19-cv-00429-CVE-JFJ (N.D. Okla.), final approval Oct. 14, 2020. Top Class Actions. topclassactions.com/lawsuit-settlements/closed-settlements/oklahoma-car-dealerships-tcpa-class-action-settlement/
- Quinn v. JK Buick GMC Inc., No. 1:23-cv-00447 (N.D. Ill.). Top Class Actions. topclassactions.com/lawsuit-settlements/closed-settlements/jk-buick-gmc-tcpa-130k-class-action-settlement/
Settlement figures are as publicly reported and reflect agreements at various stages of court approval. Settlements are not admissions or findings of liability. This guide describes LeadCentrix's own practices and is not legal advice. The field observations described here are a small sample from one market and are reported as what happened, not as a statistical study. Store names have been withheld.