Lead age tells you when a lead arrived, not whether it's alive
Open your CRM right now and look at how the follow-up task list is sorted. On most dealership installs it's sorted by day-in-process: Day 1 tasks, Day 3 tasks, Day 7, Day 14. That means a customer who opened your last three emails, clicked a payment calculator, and replied "still shopping, waiting on my tax refund" sits in the same bucket as someone who filled out a form on a vehicle you sold nine days ago and has never opened a thing.
Your rep has 40 tasks and two hours. He works top-down. Both customers get the same 90 seconds.
Age-based cadence exists because it's easy to build. It requires no judgment and no data. But it produces a predictable failure: the hottest 15% of the pipeline gets the same effort as the dead 50%, and reps quietly start skipping tasks because most of them feel pointless. Then you're managing task completion rates instead of managing deals.
The fix isn't more tasks. It's sorting the same tasks by two things you already have in the CRM: what the customer has done, and how far the deal has actually moved.
Two axes, five queues
Engagement is behavior, not opinion. Don't ask reps to rate lead temperature — they're optimistic on Monday and pessimistic on Thursday. Use signals your CRM or website tool actually logs:
- Replied to a text or email (highest-value signal there is)
- Answered a call or called in
- Opened the last two emails, or clicked a VDP/payment link
- Visited the store
- Sent a trade photo or payoff statement
- Asked about a specific stock number twice
Stage is where the deal sits in your process: no contact made, contact made/no appointment, appointment set, appointment shown/working numbers, negotiating, awaiting finance or trade decision.
Put those together and you get five queues. Build them as saved CRM searches so the rep opens one screen, not a mental filter.
Queue 1 — Live deals (work first, every day)
Appointment shown or numbers presented, any engagement in the last 72 hours, no answer yet. This is where gross lives. A rep should never start a follow-up block anywhere else.
Queue 2 — Confirmed appointments (next 48 hours)
Anyone with an appointment on the books inside two days. These are confirmation and prep touches, not sales calls. One text, one call, and a vehicle pulled up front.
Queue 3 — Engaged, no appointment
Replied, answered, or clicked in the last 10 days but hasn't agreed to come in. Highest-volume queue and the one that decides next month's traffic. Every touch here has one job: get a day and a time.
Queue 4 — Future-dated and known-reason holds
"Lease is up in March." "Waiting on the insurance check." "Wife's out of town until the 20th." These customers get pulled out of daily rotation entirely and scheduled to the date they gave you, plus one check-in halfway.
Queue 5 — No engagement
Two weeks of touches, zero replies, zero opens, no phone contact. This is where your age-based list wastes most of its effort. Low-frequency, high-leverage attempts only — different channel, different message, different time of day.
Touch cadence by queue
Cadence should be dense where there's a signal and thin where there isn't. Here's a workable structure to adapt to your store's volume:
| Queue | Frequency | Channel pattern | Exit rule |
|---|---|---|---|
| Live deals | Twice daily, 4+ hours apart | Call, then text with a specific number or answer | Sold, or customer says stop, or 5 days no reply → Queue 5 |
| Appointments (48h) | 3 touches total | Text confirm, call day-of, text on arrival window | Shown or no-show coded |
| Engaged, no appt | Daily days 1–5, then every other day to day 14 | Alternate call/text, one email with inventory | Appointment set → Queue 2, or 14 days no reply → Queue 5 |
| Future-dated | Halfway check-in + 5 days before the date | Text, then call | Date arrives → Queue 3 |
| No engagement | Day 21, 35, 60, 90 | Video text, then email, then call from a different number/rep | Any reply → Queue 3 |
The messages have to change with the queue. Cadence without content change is just harassment on a timer.
Live deal touch:
"Josh — I got the used manager to look at your Tahoe again. He moved to $23,400 on the trade, which brings you to $612. That's the number I said I'd fight for. Can you do 6:30 tonight or is tomorrow morning better?"
Engaged, no appointment:
"Dana, you looked at the silver RAV4 twice this week. It's one of two left and both have deposits pending. I can hold it under your name until 7pm tomorrow — yes or no is fine."
No engagement, day 35:
"Rachel — Mike at [store]. I'm going to stop filling up your inbox. If you already bought, congrats, tell me what you got and I'll close your file. If you didn't, reply 'still looking' and I'll send you three options under $450 a month and leave you alone otherwise."
That last one works because it gives the customer a cheap way to end the relationship, which is exactly what makes some of them re-engage instead.
Manager exceptions — where you actually intervene
Queues tell reps what to work. Exceptions tell you what to touch. Build these as three manager-only CRM views and check them once a day, ideally right after the morning save-a-deal meeting.
Exception 1: Live deal, no manager contact. Any deal in Queue 1 older than 48 hours with no manager note. Two outcomes only — you make a call or a T.O. offer today, or you and the rep agree it moves to Queue 3 and record why.
Exception 2: Engaged customer with no appointment after five attempts. This isn't a customer problem, it's an ask problem. Pull the last three texts. Nine times out of ten there's no specific day and time in any of them. Coach the ask, not the effort.
Exception 3: Queue jumps that skip stages. A deal that goes from "no contact" to "negotiating" without an appointment or a logged conversation is usually a data-entry story, not a deal. Same with anything that sat in Queue 1 for eight days — real live deals don't stay live that long.
One more worth adding at stores with more than ten reps: reassignment after 10 days of zero engagement and zero rep effort. Not as punishment. A lead with no touches in a week has an owner problem, and a second voice sometimes gets a reply the first one never would.
Rolling it out without blowing up your CRM
Do it in this order.
- Define the engagement signal list on one page. Which fields, which tools, what counts. If your site provider feeds click data into the CRM, use it. If it doesn't, engagement = replied, answered, or walked in. That's enough to start.
- Build the five saved searches. Name them by work order, not by concept: "1 — Work Now," "2 — Appointments," and so on.
- Change one thing about the rep's morning. First 60 minutes goes to Queue 1 and Queue 2 only. Nothing else gets opened.
- Kill your old aging report as a coaching tool. Keep it for accuracy checks. Stop using it to decide who calls whom.
- Review the three exception lists daily for two weeks, then decide if the thresholds are right for your volume.
The thing to watch in week three isn't task completion — that number will probably go down, and that's fine. Watch appointments set from Queue 3 and closing ratio on Queue 1. Those are the two places the sort order actually changes money.
If you want to see whether your reps' Queue 3 messages contain a specific day and time, that's the kind of thing worth pulling call and text transcripts for — MoreSignal scores exactly that ask and tells you which reps are missing it. But you can do a version of this review by hand: pull five engaged-no-appointment customers per rep on Friday and read the last three touches. You'll know inside ten minutes whether the queue is working or just renamed.