The short answer: more than half of a typical store's leads arrive outside weekday business hours, and the reply that goes out at 9am the next morning is answering a question the shopper stopped asking twelve hours ago. You don't fix this by buying a robot to talk to customers. You fix it by making unanswered leads visible — a watchdog on every lead's last inbound message, an alert to a named person, and a queue that is empty before the doors open.
Where the leads actually land
The pattern is consistent everywhere it's been measured. A widely-circulated 30-day industry study of 7,041 leads across 50 dealerships found 53% arrived outside 9am–6pm on weekdays — roughly 75 after-hours leads per store, per month — and that 3,743 of the leads in the sample were never answered at all.
Two honest caveats before you build a plan on that. First, it's vendor-published research, and vendor research tends to find the problem the vendor sells against — treat the shape as real and the decimal places as marketing. Second, you'll see that study quoted alongside a "$400,000 of annualized gross exposure per rooftop" figure. That number is an estimate built by multiplying unanswered leads by an assumed close rate and an assumed gross. We're not going to repeat it as a fact, and neither should the person pitching it to you. Run the multiplication on your own store's close rate and your own gross — the answer will still be large enough to act on.
| What the data shows | Rough figure | What it means on your floor |
|---|---|---|
| Leads arriving outside weekday 9–6 | ~53% | Your busiest "salesperson" is the overnight queue |
| After-hours leads per store, per month | ~75 | Two to three every single night |
| Never answered at all (sample of 7,041) | 3,743 | Over half went unanswered — not slow, unanswered |
| Odds of qualifying at 30 min vs. 5 min | ~21× worse | A 12-hour gap isn't "late," it's a different outcome |
That last row is from the lead-response research covered in our lead response time guide — and it's the reason after-hours matters more than the raw percentage suggests. If the odds collapse inside the first thirty minutes, an overnight lead isn't answered late. It's answered in a different universe.
"After hours" is bigger than nights
Most stores picture 2am when they hear this. The real gaps are more ordinary, and they add up faster:
- Evenings, 6pm to 10pm. The single densest window for form fills. People shop for cars after dinner, on the couch, on their phone.
- Sunday. Closed in most markets, and one of the heaviest browsing days of the week. A Sunday-afternoon lead answered Monday at 9am has had a nineteen-hour head start on you — and your competitor two towns over may not be closed.
- Saturday on the floor. Technically open, functionally unavailable. Every rep is with a customer, and the inbound queue is nobody's job until the last delivery is done.
- The shift seam. Late afternoon, when the closing rep assumes the opener handled it and the opener assumed the closer would.
- Holidays and long weekends. Three days of accumulation, and the pile is triaged newest-first on Tuesday — which means the oldest and coldest leads are the ones that get skipped.
Notice what these have in common: none of them are a technology problem. They're an ownership problem. A lead can be assigned, sitting in the CRM, visible to five people, and still be nobody's responsibility until someone happens to refresh a screen.
Why "just buy a bot" is the wrong first move
Search this topic and you'll find a wall of AI answering services promising to talk to your customers overnight. Some of them work reasonably well at booking a simple appointment. But leading with one is backwards, for three practical reasons.
It automates a process you haven't defined. If nobody owns unanswered leads at 9am, adding a machine at 9pm just moves the confusion. You'll get a transcript of a conversation nobody follows up on.
A car deal is not a support ticket. The overnight questions that matter are trade values, payments, whether a specific unit is still available, and whether financing is realistic. Those are the questions where a wrong-but-confident automated answer costs you the deal and, occasionally, costs you a complaint. The safe automated answer is so hedged it reads like a brush-off.
It hides the metric you actually need. Once a bot "responds" to everything, your response-time report goes green and your close rate doesn't move. You've deleted the symptom and kept the disease.
There's a version of automation that's genuinely worth it — but it points at your staff, not your customers.
| Answering bot | Alert + accountability | |
|---|---|---|
| Who talks to the customer | Software | A person, informed |
| What it optimizes | Response sent | Response that closes |
| Failure mode | A confident wrong answer about price, payment or availability | A manager's phone buzzes and gets ignored — visible, and fixable |
| Effect on reporting | Response time looks solved | Shows exactly which leads, hours and reps leak |
| Real trade or finance question | Handled poorly | Handled, because a human handles it |
| Cost of being wrong | A customer misinformed by "you" | A few minutes |
To be direct about our own position: GhostDrive does not respond to your customers. It watches, flags, and escalates. The reply is always a human's, which is also why it can be a good one.
What accountability actually looks like
Four pieces, in order. None of them require you to staff the store at midnight.
1. A watchdog on the last inbound message
The signal that matters isn't "a lead arrived." It's a lead whose most recent message is inbound and unanswered. That single condition catches the overnight form fill, the 8pm text reply on an active deal, and the Saturday voicemail nobody returned — all of which look completely different in a CRM and identical to the customer.
2. Thresholds that change by hour
One SLA for all 24 hours is a rule people learn to ignore. Set bands you'll actually honour: 15–30 minutes during open hours, a longer overnight window with a hard "cleared before open" rule, and a shorter, sharper threshold for high-intent signals — a finance application, a specific-VIN question, a returning shopper who's been on the same unit three times this week.
3. Escalation to a name, not an inbox
An alert to a shared inbox is a rounding error. It has to reach a specific manager, on the device they already have in their hand, in the channel they already live in — and it has to say which lead, from which source, how long it's been sitting, and what the customer actually said. "You have 3 unanswered leads" is noise. "Sarah M., 8:42pm, asking if the silver Highlander is still available — unanswered 11 hours" is a task.
4. A morning queue with a defined end state
The first fifteen minutes of the day belong to last night. Oldest-first, not newest-first, because the oldest is the one closest to gone. The queue's success condition is empty before the doors open, and the person who owns "empty" is a person, not a policy.
And when the rep does call back, they should already know what the shopper looked at, whether they applied, and what they're trading — so the first human contact is the right answer, not a stall. Twelve hours late and uninformed is how a slow lead becomes a lost one.
Measure four numbers, not one
"Response time" as a single store-wide average will hide this problem completely — a fast weekday afternoon buries a dead weekend. Split it:
- After-hours share of leads. What percentage arrived outside your open hours, by source. This is the size of the problem, and it's usually bigger than the desk believes. It also tells you which channels are buying you overnight demand you aren't staffed for — the cost of a third-party lead doesn't change based on whether you answered it.
- First-response time, banded by hour of arrival. Same metric, split into open / evening / overnight / weekend. The bands are where the story is.
- Unanswered count at open. How many leads are sitting with an inbound last message when the doors unlock. Target zero. Trend it daily.
- Close rate by arrival window. Do 8pm leads close worse than 2pm leads at your store? If yes, by how much — that gap, times your gross, is your own honest version of the exposure number, and it's the one you can defend in a meeting.
All four belong on the same daily board as your other numbers — see the dealership KPI dashboard guide for where they sit. And they only work if lead source is trustworthy in the first place, which is an attribution problem before it's a staffing one.
Building the after-hours shift you already have
You don't need a night crew. You need three decisions made once:
- Name the on-call. One manager per evening, rotating. Their job isn't to work deals at 10pm — it's to triage: reply to the two that are genuinely live, and leave the rest queued with a note.
- Decide what's worth waking up for. Most stores land on: a finance application, a specific-unit availability question, and any reply on an active deal. Everything else waits for morning, deliberately — which is different from waiting by accident.
- Give evenings a real reply, not a stall. A short, human, honest message at 8:50pm — even "I'll have that trade number for you first thing, and the Highlander is still here" — beats an eloquent 9:15am template. Being present beats being polished.
The point isn't heroics. It's that the store knows what's sitting there, and one named human is accountable for it. That's a system, and systems survive vacations.
The checklist
Whatever you use, require all six:
- Every source's leads land on one board — website, marketplaces, social, phone — deduped to one record per shopper
- The system flags any lead whose last message is inbound and unanswered, not just "unassigned"
- Response thresholds differ by hour band, and overnight has a hard clear-before-open rule
- Alerts reach a named person on a device they carry, with the customer's actual question in the alert
- The rep gets the shopper's history before they call back, not after
- After-hours share, banded response time, unanswered-at-open and close-rate-by-window are reported and reviewed weekly
Questions dealers ask
- What percentage of car dealership leads come in after hours?
- A widely-cited 30-day study of 7,041 leads across 50 dealerships found about 53% arrived outside 9am-6pm on weekdays - roughly 75 after-hours leads per store per month. It is vendor-published research, so treat the precise figure with care, but the shape holds up across stores: evenings and Sundays are among the heaviest shopping windows, and they are exactly when the desk is empty. The number worth trusting is your own - split your last 90 days of leads by hour of arrival and you will usually find somewhere between 40% and 60% landed outside open hours.
- Should a dealership use an AI bot to answer leads overnight?
- It can help with simple appointment-setting, but it should not be the first move. Overnight questions in car sales are mostly about trade value, payments, availability and financing - the places where a confident wrong answer costs you the deal. Worse, once something replies to everything, your response-time report looks solved while your close rate does not move. Define ownership first: detect the unanswered lead, alert a named manager, and let a human give the answer. Automate the alert, not the conversation.
- How do you track unanswered leads at a dealership?
- Watch the direction of the last message, not the assignment status. A lead can be assigned to a rep, visible in the CRM, and still be unanswered - so the condition to monitor is that the most recent message is inbound and no reply has gone out. Set a time threshold that varies by hour of day, escalate to a specific manager by name when a lead crosses it, and report an unanswered-at-open count every morning with a target of zero. Assignment is not an answer, and a shared inbox is not an owner.
