Can AI follow up on the work my customers declined?
Yes. Auto Advisor's Retention agent drafts the check-in and Google-review ask after every closed repair order, chases declined work while it still matters, and nudges quiet customers to rebook. Every message is a draft a human sends from your own tools; the product never messages your customers directly.
The most common repair-order band for general independent shops is $500 to $749, so every declined job you recover is roughly another full RO. (PartsTech 2025 survey of 752 U.S. shops.)
What does the Retention agent actually do?
Brings customers back after the visit: the review ask, the declined-work chase, and the maintenance nudge, every message a draft you send yourself.
- 01
Drafts a check-in plus Google-review request for every repair order closed one to ten days ago.
- 02
Drafts a declined-services chase while the RO is under 60 days old, re-drafting at most once per 14-day window, with the dollar figure summed from stored declined prices, never invented by the AI.
- 03
Drafts a maintenance reminder for customers with no closed RO in 150+ days, capped at one per month.
- 04
Picks the channel (email if on file, otherwise SMS) and honestly skips and counts customers with no contact info.
What it reads to do the work
- Your closed repair orders, declined-work items, vehicles, and customer contact info
- The 70% retention benchmark, stamped on each draft for context
- The autonomy mode you set for it (Off is fully read-only)
How do I stay in control of what it does?
Every agent runs in the mode you choose, Off, Approve, or Auto, and every agent starts in Approve so a human signs off before anything happens. Here is what Retention can do, with the gate on each action, and the rails that hold in code, not in a policy document.
The actions, and their gates
- Queue a review-request draft you send yourselfYou approve it
- Queue a declined-work chase draft with the real dollars at stakeYou approve it
- Queue a maintenance-reminder draftYou approve it
The guardrails, held in code
- Draft-only in EVERY mode, including Auto, enforced in the action-policy code, not just convention.
- Anti-spam pacing built in: one chase per 14-day window, one reminder per month, with per-window idempotency so re-runs queue nothing new.
- A cap of eight drafts per run.
- Customer names and job descriptions are sanitized and fenced as untrusted data; the AI may never set a dollar figure.
What does Retention not do yet?
Every claim on this page maps to shipped, tested code, and the same discipline applies to the limits. Here is where the edge of the product is today, stated plainly, so nothing on your shop rests on a promise.
- You send every message from your own phone or email tool, which also means final TCPA and Do-Not-Call screening happens at your send step; the product does not screen numbers against a DNC list for you.
- It runs when triggered from the dashboard, not yet on an always-on cadence timer.
- Whether a chased job actually came back is not yet measured; the queue shows what was drafted and what you marked handled.
Which numbers does Retention actually move?
Only measured numbers are listed here; a metric the product does not compute does not get claimed.
- Customer retention rate versus the 70% benchmark (computed)
- Declined-work dollars put back on the table, per chase draft
- Review requests drafted per run
The rest of the crew
Retention is one of fifteen agents. Each has its own charter page, its own autonomy dial, and the same honesty rule.
Put Retention to work in your shop.
Request a Service-Drive Audit. We map exactly where the crew recovers time and revenue, on your numbers, in a live demo on sample data.
No obligation · no high-pressure pitch · plain, public pricing