Review Automation
The Delivery-Day Window: A 5-Step Review System for Auto Dealerships
Most dealerships ask for reviews at the worst possible moment and from the wrong half of their customers. This is a five-step system for capturing reviews from both the showroom and the service drive, on a schedule that actually holds.
by Jerrod Anthraper
It is 7:40 on a Saturday night. The finance office ran long, the customer has been at your store for four hours, her kid is melting down in the lounge, and she just wants to drive her new Highlander home. Your salesperson hands her the keys, then slides a card across the desk and says, "Hey, if you get a chance, we would really appreciate a five-star Google review." She says of course. She means it. That review is never getting written, and everyone involved already knows it.
The leak: your review rate is set by your worst moment
Dealerships do not have a review problem so much as a timing problem. The ask happens at the single most depleted point in the entire customer experience — the end of a long transaction, when the customer wants out of the building.
The result is a review profile built almost entirely out of two groups: the handful of customers who genuinely love their salesperson, and the customers who are angry. Nobody in the middle ever writes anything. That skew matters because about 92% of consumers read reviews before buying, and for a purchase this size they are not skimming — they are reading recent ones, looking specifically for how your store behaves when something goes wrong. A thin, lopsided review profile does more damage at a dealership than at almost any other kind of local business.
Why it happens: the ask is attached to the wrong event
The root cause is that most stores tie the review request to the delivery, because delivery is the moment the salesperson controls.
But delivery is not when the customer is happy. Delivery is when the customer is finished. The happiness arrives later — the first Monday commute, the first time the adaptive cruise does something impressive, the first tank of gas that cost less than the trade-in used to. That is when someone will actually write four sentences about your store, and by then nobody has asked.
The second cause is that the service drive, which touches the same customer eight or ten times over the ownership cycle, is usually excluded from the review program entirely. Sales owns reviews. Service owns CSI. The two systems do not talk, so the department with the most frequent customer contact contributes almost nothing to the store rating that sells cars.
The five-step delivery-day window system
This is built to run on a schedule rather than on whoever remembers.
1. Move the ask 48 to 72 hours past delivery
Stop asking in the finance office. Set the request to fire two to three days after the customer takes the car home — after the first commute, before the experience fades.
Keep the delivery-day mention, but change what it is. Instead of asking for the review, the salesperson says, "I am going to check in with you in a couple of days to make sure everything makes sense with the tech — and if we did right by you, I will ask you to say so publicly then." That reframes the follow-up as expected rather than as an interruption, which is most of the battle.
2. Ask a question, not for a rating
"Please leave us a five-star review" produces either nothing or four generic words. Ask something specific enough that the answer is already a review.
For sales: "What was the one thing that made you decide to buy from us instead of the other store you looked at?" For service: "Was the price you were quoted the price you paid?" People can answer a specific question in twenty seconds. They cannot answer "leave us a review" without composing an essay first, so they close the tab.
3. Put the service drive on the same system
Your service department is your review engine and it is almost certainly idle. A sales customer generates one review opportunity every four or five years. A service customer generates one every few months.
Set the service request to fire the day after the repair order closes, not at pickup, and route it by advisor so you can see which advisors generate reviews and which generate silence. The advisor-level data is worth as much as the reviews themselves — it will tell you within two months exactly where in your drive the customer experience actually breaks down.
4. Route friction before it becomes a one-star
Build a single intermediate step: ask how the experience went before asking for the public review. If the answer is positive, send the review link. If it is negative, route it to a manager the same day.
Be careful about how you frame this internally. This is not about suppressing bad reviews, and any system sold to you on that basis is a liability. It is about giving an unhappy customer a faster path to a human than to a keyboard. A store that resolves the complaint on day one usually earns the review anyway, and it earns the more valuable kind — the one where a customer describes a problem being fixed.
5. Set a weekly floor and hold one person to it
Pick a number of reviews per week that your store must produce, write it on the same board as your sales and service targets, and give it an owner with a name.
Most stores that install steps one through four still fail here, because a review program with no owner degrades to zero within about six weeks. The floor should be modest enough to be real — start at five or ten a week — and reviewed every Monday alongside everything else you measure. Volume and recency matter as much as the average. A 4.6 built from forty reviews in the last quarter outperforms a 4.8 built from eleven reviews written three years ago, because the customer reading them is checking dates.
What the rating is worth in dollars
The reason to run this as a system rather than a campaign is that the rating itself moves conversion on every other channel you pay for.
Moving from a 3.5-star average to a 4.5-star average can lift conversion by as much as 270%, and each additional star correlates with a 5% to 9% increase in revenue.
For a dealership, that lift applies to traffic you have already paid for — the third-party lead, the paid search click, the customer who saw your inventory on a listing site and then searched your store name before deciding whether to drive over. You are not buying more traffic. You are keeping more of what you already bought.
Automating the part that always slips
Every step above is straightforward, and step five is where nearly every store fails, because it depends on a person remembering to do a low-urgency task during a week that is full of urgent ones.
That is the piece James, our AI sales agent, is built to carry. He handles the timed follow-up after delivery and after the repair order closes, asks the specific question rather than begging for stars, routes the unhappy responses to a manager the same day, and keeps doing it in the weeks when the store is slammed. Our transportation client AMG Transport Co. went from five or six reviews a month to five or six a day after putting this on automation — the same volume shift a dealership service drive is capable of producing, and for the same reason: the asks kept happening after the humans got busy.
Before you change anything, pull your review counts by month for the last twelve months and split them into sales and service. If service is contributing under a quarter of your total, you already know which of the five steps to start with.