How to Build a Dealership Follow-Up Process That Actually Runs

What Does a Follow-Up Process That Actually Runs Require?

A follow-up process that runs requires four things working together: defined trigger events that start the clock, a timing window capped at 24 hours, assigned ownership at every handoff, and a contact rate measurement system built from real DMS data. Miss any one of the four and the process degrades into whatever the BDC gets to that day.

Most dealerships have a version of two or three of these. They know roughly what should trigger a call. They have a rough sense of who's supposed to make it. What's almost always missing is the measurement layer - the part that tells a GM whether the process is actually running or just exists on paper. Without it, "the BDC handles it" is unfalsifiable. Nobody can say whether it's true.

The rest of this piece walks through each of the four components in the order a dealer should build them: triggers first, because nothing else matters if the clock never starts; measurement second, because you can't fix what you haven't measured; capacity third, because measurement usually reveals a staffing math problem; and the AI-human handoff last, because it's the mechanism that makes the other three sustainable at volume.

What Are the Four Trigger Events Every Follow-Up Process Needs?

Four events should start a follow-up clock at every dealership: a repair order closing, a showroom visitor leaving without a deal, an internet lead submission, and a service appointment reminder window opening. Each has a different timing requirement, and treating them identically is one of the most common reasons follow-up processes fail.

Service RO close: the clock starts the moment the repair order closes in the DMS - CDK, VinSolutions, Dealertrack, or TEKION all timestamp this event. The call should go out within 24 hours, and ideally within a few hours if the visit included any flagged friction. This is also the trigger tied most directly to CSI survey timing, since OEM surveys typically go out 3-10 days after the visit.

Unsold showroom visit: the desklog entry created when a visitor leaves without buying is the trigger. The window here is tighter than most dealers treat it - contact rates and conversion both drop sharply after 72 hours, and NADA data shows buyers overwhelmingly choose the first dealer to follow up, not the one they originally preferred.

Inbound internet lead: the trigger is the form submission or chat handoff itself, and the timing requirement is the strictest of the four. Response time is the single biggest variable in whether an internet lead converts, and a growing share of dealers now respond within 15 minutes - which resets what "fast" means for everyone still working off a next-morning callback list.

Service appointment reminder: this trigger runs in the opposite direction - it fires before the visit, not after, confirming the appointment and reducing no-shows. It's the trigger dealers most often skip building a process around, even though a missed appointment is lost service revenue with no recovery step behind it.

“Treating all four trigger events with the same timing rule is one of the most common reasons follow-up processes fail.”

How Do You Calculate Your Dealership's Actual Contact Rate?

Most dealers don't know their actual contact rate because their CRM dashboard counts voicemails, unanswered attempts, and texts sent as "contacts" alongside live two-way conversations. Correcting for this is the single most useful thing a GM can do before touching anything else in the process.

Start with the denominator: pull your total trigger-event count for the past 30 days from your DMS - closed ROs, unsold desklogs, internet leads, whatever population you're measuring. That's every customer who should have received a follow-up attempt.

Then pull the numerator correctly. From your CRM, isolate dispositions logged as live conversations only - not "attempted," not "voicemail left," not "text sent." If your CRM's disposition codes don't cleanly separate these, pull a sample of 50 call recordings and audit the split by hand. The gap between what your CRM reports and what actually happened is usually larger than dealers expect.

Divide live conversations by total trigger events. For most stores running this calculation correctly for the first time, the number lands below 25% - consistent with the 15-20% industry average for manual BDC outreach. We've written a full breakdown of contact rate benchmarks by use case if you want to see how your number compares to the top quartile.

How Many Outbound Contacts Can One BDC Rep Actually Handle?

A single BDC rep tops out around 100-150 outbound sales-lead attempts per month before call quality and follow-through start to drop (Strolid, 2024). That ceiling is the math most dealers skip when they assume their existing team can cover a growing trigger-event volume.

Run the arithmetic for a typical store: 400 unsold desklogs a month, plus 300 closed ROs worth a follow-up call, plus whatever internet lead volume comes in. That's 700+ trigger events competing for a rep capacity of 100-150 attempts each. A team needs 3-4 dedicated agents just for outbound coverage - before any of them touch an inbound call, a warm-lead conversation, or an appointment reschedule.

This is where the capacity gap actually opens. It's not that BDC reps are underperforming individually. It's that the volume of trigger events at a mid-size store structurally exceeds what a team of two or three people can physically dial through in a month, regardless of how skilled they are.

“It's not that BDC reps are underperforming. It's that trigger-event volume structurally exceeds what two or three people can dial through in a month.”

Is BDC Capacity or Process the Real Constraint at Your Store?

If your contact rate is low but your team is at or near the 100-150 attempts-per-rep ceiling, capacity is the constraint - no amount of process tightening fixes a math problem. If your team is well under that ceiling and contact rate is still low, the constraint is process: unclear ownership, no defined timing window, or trigger events that aren't reliably reaching the BDC queue in the first place.

The diagnostic is simple. Pull your team's logged attempt volume for the past 30 days and divide by headcount. If each rep is already near 150 attempts, adding process discipline won't move the contact rate number - the team is out of hours, not out of effort. If reps are logging 60-80 attempts a month with room to spare, something upstream is broken: leads aren't routing fast enough, ownership isn't assigned clearly at handoff, or the timing window isn't enforced.

Most dealerships that run this diagnostic honestly find it's both. A capacity shortfall usually starts as a process problem - some trigger events never make it into the queue at all - and gets worse as volume grows past what the existing headcount can absorb, even with perfect process discipline.

How Should the Handoff Between AI First-Contact and Human Escalation Work?

The handoff that works in practice: AI owns first-contact attempts on every trigger event - every closed RO, every unsold desklog, every internet lead gets an outbound call inside the 24-hour window, regardless of volume. AI's only job at that stage is determining who's still in the market or who's a CSI risk, then routing them immediately to a human.

Human BDC owns everything past that first contact - negotiation, scheduling, objection handling, and any escalation flagged as a detractor or a genuinely upset customer. This is the same coverage-versus-complexity split that separates AI BDC from traditional BDC: AI solves the problem of physically reaching everyone, humans solve the problem of what happens once someone's actually on the line.

The warm transfer is the part dealers get wrong most often. A handoff that dumps a transcript into a CRM task queue with no urgency flag is barely better than no handoff at all - the human rep still has to figure out who's actually worth calling back first. A handoff that tags severity (routine follow-up vs. flagged detractor vs. ready-to-buy) and notifies the right person immediately is what makes AI first-contact and human escalation function as one process instead of two disconnected systems.

What Metrics Should Your Follow-Up Dashboard Actually Track?

Five metrics separate a follow-up process that's measured from one that's guessed at: contact rate, first-contact time, escalation rate, resolution rate, and conversion rate by follow-up timing bucket. Most dealer CRMs can produce all five - the issue is almost never data availability, it's that nobody built the report.

Contact rate: live conversations divided by total trigger events, calculated separately by trigger type (RO close, desklog, internet lead) since each has a different baseline.

First-contact time: the elapsed time between the trigger event firing and the first attempt being made. This should be tracked in hours, not days - a report that only shows same-day vs. next-day hides the difference between a 40-minute response and a 22-hour one.

Escalation rate: the percentage of contacted customers flagged for human follow-up, whether for a CSI risk, a complex objection, or a ready-to-buy signal. A rate that's consistently near zero usually means the flagging criteria are too narrow, not that customers have nothing to escalate.

Resolution rate: of escalated customers, what percentage received a human follow-up and reached a documented outcome - resolved, scheduled, or closed - rather than sitting in a queue.

Conversion rate by follow-up timing bucket: conversion outcomes segmented by how fast the first contact happened (under 1 hour, 1-24 hours, 24-72 hours, 72+ hours). This is the metric that proves or disproves the value of tightening your timing window, because it shows the actual revenue difference rather than an assumed one.

Frequently Asked Questions About Building a Dealership Follow-Up Process

What's the single biggest mistake dealers make building a follow-up process? Treating all trigger events with the same timing rule. A closed RO, an unsold desklog, and an internet lead each carry a different urgency, and a process that batches them into one daily call list misses the window on the fastest-decaying ones - internet leads and unsold showroom traffic.

How do I know if my follow-up process is actually running or just exists on paper? Pull your true contact rate - live conversations divided by total trigger events, not attempts. If you can't produce that number from your CRM in under ten minutes, the process isn't measured, which usually means it isn't running consistently either.

Should I hire more BDC staff or add AI first-contact coverage? Run the capacity math first: a rep tops out around 100-150 outbound attempts a month. If your team is already near that ceiling and contact rate is still low, adding headcount or AI coverage solves the volume problem. If your team has capacity to spare and contact rate is still low, the fix is process, not more people.

What CRM or DMS systems does this process work with? The four-trigger model works with any DMS that timestamps RO closes, desklog entries, and lead submissions - CDK, VinSolutions, Dealertrack, and TEKION all support this. The process itself is DMS-agnostic; what matters is that trigger events reliably reach whoever (or whatever) is making the first call.

How long does it take to stand up a working follow-up process from scratch? Defining the four triggers and timing windows can happen in a week. The harder part is the measurement layer - getting contact rate reporting built correctly usually takes another few weeks of CRM disposition cleanup. Adding AI first-contact coverage on top, once DMS data is flowing, typically goes live within days.