Most rooftops do not have a "lead problem." They have a time allocation problem that looks like a lead problem in the month-end report.
The BDC is buried in unfinished CRM tasks, unfinished internet leads from last night, and phone tag on units that already sold. Marketing buys more traffic. The queue gets longer. Close rate dips. Someone schedules a demo for another point solution.
I sell cars for a living and I build dealership automation on the side. This post is the audit I want every GSM and BDC manager to run before they buy another inbox.
Related: After-hours dealership leads — read that for the overnight gap. This one is the daytime clock.
The 5-bucket BDC week
Have each BDC rep (or the working manager) log tasks in 15-minute blocks for five business days. No fancy software — a shared sheet is enough.
| Bucket | Examples | Why it matters |
|---|---|---|
| A. Live conversations | Inbound calls, booked appointments, solid chat handoffs | Revenue work |
| B. Speed-to-lead on fresh internet | First response to web/chat/marketplace leads < 5–15 min | Decay curve is brutal |
| C. CRM hygiene | Correcting statuses, dead dupes, sold-unit cleanup, forced tasks | Often 20–40% of the day when CRM is dirty |
| D. After-hours catch-up | Working the overnight form dump from 8 PM–9 AM | Symptom of no night coverage |
| E. Internal coordination | Desk-manager scavenger hunts, "is this VIN still here?", CSI fire drills | Coordination tax |
You are looking for two ratios:
- (A + B) / total — time in true selling motions.
- (C + D) / total — time in preventable sludge.
If sludge is over ~30% of the week, more lead vendors will not save you. You will just drown faster.
CRM hygiene checklist (do this before AI)
AI on top of a filthy CRM creates confident wrong answers. Clean these first:
- Sold / wholesale / dealer-trade status must match the lot within 24 hours.
- One lead = one record. Merge obvious dupes weekly.
- Source fields honest (web, phone, chat, 3rd party) — you cannot optimize what you mis-tag.
- Next task dates realistic. A task pile 400 deep is not a process; it is guilt.
- Phone + email required on internet leads before they enter long-term nurture fiction.
After-hours tools that quote live inventory only help if inventory feeds are trustworthy. Garbage in, awkward chat out.
After-hours is a separate line item
In the audit, split D from everything else. Overnight form dumps are not a BDC motivation issue. They are a coverage design issue.
What night shoppers ask (from real CRM notes and chat logs):
- Is this VIN still available?
- Out-the-door range with X down?
- Color / package / bed length on the lot?
- Test drive tomorrow?
- Service appointment Saturday?
Those are answerable from live inventory + calendar rules. They should not all become 9:05 AM phone debt.
Our dealership chatbot product is priced for that job: $749/mo for live-inventory chat (own instance, own branding), $749/mo for a custom dealer site, $1,498/mo for both — no per-conversation meter. Details: /dealerships/ · /product-pages/dealership-chatbot.html.
I am not telling you to fire the BDC. I am telling you to stop making humans retype "yes, that VIN is here" 40 times a night in the morning.
What the chatbot should own vs. what stays human
| Task | Chat / automation | Human BDC |
|---|---|---|
| VIN availability, basic specs, hours, directions | Yes | Exception handling |
| Payment range / payment estimator with disclosed assumptions | Careful yes | Desking hard numbers |
| Route to sales vs service vs parts | Yes | Complex escalations |
| Trade appraisal final | No | Yes |
| Credit decisioning | No | Yes |
| Angry CSI / deal-jacket problems | No | Yes |
| Appointment set on calendar | Yes with rules | Confirm high-gross appointments |
If a vendor demo starts with "it replaces your BDC," leave the room. If it starts with "it clears A and D so your people live in A and hard deals," keep listening.
Half-day audit agenda (copy/paste)
09:00–09:30 — Pull last 14 days: lead volume by hour, by source, by response time.
09:30–10:30 — Shadow or record two BDC hours. Tag every task A–E.
10:30–11:00 — Inventory feed check: 10 random VINs on the site vs. DMS/lot.
11:00–11:30 — Overnight sample: 20 leads that arrived 8 PM–8 AM. Time to first human response. Outcome.
11:30–12:00 — Decide one fix only for the next 14 days:
- Hygiene sprint, or
- After-hours inventory chat pilot, or
- Kill one low-quality lead source.
One fix. Not five.
Numbers worth writing on the whiteboard
Use your data. These are prompts, not universal KPIs:
- Median first response on web leads (daytime vs overnight).
- % of overnight leads touched before 10 AM.
- % of BDC time in bucket C (hygiene).
- Appointments set per BDC labor hour.
- Sold-unit / wrong-status complaints from shoppers (qualitative count for two weeks).
When overnight response collapses and daytime hygiene is high, buy coverage for the night clock — not another coat of paint on the website.
Honest constraints
- No day-job or partner rooftops named as customers here. Day-job observation is not a case study logo.
- $749 / $749 / $1,498 pricing is current as of 2026-08-08 on our dealership surfaces. If you are reading a printout later, check the live page.
- Chat does not fix a broken desk process or a manager who will not enforce CRM statuses.
Bottom line
Run the time audit. Separate selling minutes from sludge minutes. Clean the inventory truth. Then cover the hours your humans are asleep with a tool that reads live stock, not a brochure bot.
If you want a second set of eyes on the spreadsheet, send the anonymized hourly mix — source counts only, no customer PII.
Talk to Peter about a dealership after-hours + BDC workflow review.