Index
Why checking satisfaction first changes the outcome
The right time to ask for the review
The two-question flow
How AI personalises the message
What to do with each type of reply
How to manage the channels: WhatsApp, email and phone call
Replying to reviews with the help of AI
What to measure beyond the average rating
Learning from the language of reviews
More than 750 dealerships already use Dealcar for their daily operations
Frequently asked questions
Why checking satisfaction first changes the outcome
Asking someone who had a bad experience for a review has two negative consequences. The first is obvious: it can generate a negative review that damages the dealership's reputation. The second is less obvious but just as important: the unhappy buyer who gets the review request before anyone has asked whether everything is OK feels that the dealership only remembers them when it wants something.
The right flow reverses the order: first ask how everything is going, then ask for the review only if the answer is positive. That order does two things at once: it detects problems that can be fixed before they escalate into a public review, and it only asks for feedback from buyers who are in a position to give a positive, honest rating.
AI adds value to this process at three points: it automates the first satisfaction message so it arrives at the right time without anyone having to remember, it classifies the reply (positive, neutral, issue) and decides the next step based on that classification, and it personalises the message with the real data from the deal (name, car model, salesperson's name, site).
The right time to ask for the review
Timing matters as much as the message. A message sent too early (on the day of handover) arrives before the buyer has had time to use the car and form a real opinion. A message sent too late (three weeks later) arrives when the memory of the experience has already faded.
The most effective window for the first satisfaction message is between 24 and 72 hours after the vehicle is handed over. By then the buyer has driven the car, has had time to check the paperwork and can give an informed assessment.
See also how to manage reservations and handovers at a dealership.
There are situations where you need to wait longer or not ask at all. If there is an open issue (missing paperwork, a mechanical problem found in the first week, a complaint in progress), the satisfaction message isn't sent until the issue is resolved. Asking for a review in the middle of an open complaint is one of the mistakes that most damages the relationship with the buyer.
There are also moments when it makes sense to ask for the review outside the standard after-sales cycle: after an issue has been resolved properly (a buyer who was unhappy and ended up satisfied with the resolution is a very good candidate for a review), or after a spontaneous positive signal on WhatsApp or during a call.
Timing | Ask for a review? | Recommended action |
|---|---|---|
Order signed | No | Thank them for their trust |
24-72h after handover with no issues | Yes, if satisfaction is positive | Check satisfaction first |
Open issue | No | Resolve first |
Issue resolved properly | May be appropriate | Ask about the resolution |
Weeks later with no prior contact | Only with context | Experience survey |
The two-question flow
The most effective flow has two sequential steps, not one.
The first step is the satisfaction question: "Hi, [name]. [X days] ago you picked up your [model] from us. Is everything OK with the car and the handover?". That question is open-ended; it doesn't lead to a numerical rating or a link. It just asks how things are going.
The second step depends on the reply. If the reply is positive ("all perfect, the car is great"), the system thanks them and offers the direct link to the dealership's Google profile: "We're really glad to hear that. If you have a minute, it would help us a lot if you left your review on Google. Here's the direct link: [link]". If the reply points to a problem ("the second key is missing", "the satnav doesn't work", "I still haven't received the paperwork"), the system doesn't ask for the review: it opens an issue in the CRM with a description of the problem and assigns an urgent task to the person responsible.
See also how to automate WhatsApp Business in a dealership with AI.
That second step is where AI adds the most value: it classifies the reply and automatically decides whether the next step is the review link or opening an issue, without anyone having to read each reply manually.
How AI personalises the message
Personalisation is what separates a review request that reads as genuine communication from one that reads as automated spam.
The elements the system can personalise automatically from the sales file data are: the buyer's name, the exact model and trim of the vehicle handed over, the name of the salesperson who handled the deal, the site where the handover took place and the number of days since handover.
With those elements, the message "Hi, Carlos. Three days ago you picked up your Volkswagen Tiguan R-Line with Ana at our Alcobendas centre. Is everything OK with the car and the handover?" has a level of personalisation that shows the dealership knows exactly who that buyer is and what they bought.
What AI shouldn't do is write the text of a review for the buyer to copy, or make any kind of benefit or discount conditional on receiving a positive rating. Those practices breach Google's policies and produce reviews that don't reflect the real experience.
Read also how the AI sales agent works in a dealership.
What to do with each type of reply
Replies to the satisfaction message fall into four categories, each with a different next step.
The promoter buyer replies in a clearly positive way: "Delighted, all perfect, great service". The system thanks them and sends the direct link to the Google profile. If they haven't left the review within the next 3 or 4 days, a single, discreet reminder can be sent.
The neutral buyer replies ambiguously or without an explicit assessment: "Yes, the car's fine". The system can ask a follow-up question to better understand the experience before deciding whether to send the review link.
The buyer with an issue mentions a specific problem. The system opens the issue in the CRM with a summary of the problem, assigns a task to the person responsible with a priority based on how urgent the problem is, and doesn't send the review link. Once the issue is resolved, a new message can be sent asking whether everything was sorted out properly, and at that point, if the reply is positive, the review can be raised.
The buyer who doesn't reply. If there is no reply within 48 or 72 hours, a reminder of the satisfaction message can be sent, but only one. After that, don't push.
How to manage the channels: WhatsApp, email and phone call
Each channel has a different strength for review requests.
WhatsApp is the channel with the highest open rate and the most direct way to send the Google link with a single tap. Its advantage is that the buyer can leave the review straight from their phone with no extra steps. It's the recommended channel for most buyers under 55.
Email allows more context and is appropriate when the dealership uses the buyer's email as its usual communication channel. The open rate is lower than WhatsApp, but the message can include more contextual information and the review link can be more prominent.
A phone call lets you pick up on nuances that text doesn't capture: the buyer's tone, a doubt they wouldn't have expressed in writing, a complaint they wouldn't have sent by text. It's the most effective channel for older buyers or for high-value deals where the personal relationship matters more. The downside is that it takes up the team's time and the buyer has to open the review link later from another device.
Replying to reviews with the help of AI
Review management doesn't end when the rating comes in: the dealership's reply to each review is part of the business's reputation.
AI can prepare personalised draft replies for each review, adapting the tone depending on whether it's positive, neutral or negative, and mentioning specific details from the review (the model bought, the salesperson mentioned, the situation described). Those drafts save the person responsible time, as they can review and publish instead of writing from scratch.
What always has to go through human review is the reply to negative reviews or to reviews that mention specific situations where the facts need checking. A reply to a negative review that gives incorrect information or comes across as defensive can make the problem worse instead of solving it.
Replies that sound like a template (same structure, same phrases on every review) come across as automated and reduce the reply's credibility. A buyer who left a detailed positive review deserves a reply that mentions something specific they wrote, not a generic thank-you.
See also how the AI marketing agent uses the dealership's content.
What to measure beyond the average rating
The average Google rating is an outcome indicator, not a process indicator. It says nothing about why buyers are satisfied or dissatisfied, or about what is getting better or worse.
The six metrics you need to keep visible to manage reputation operationally are: requests sent vs reviews received (the process conversion rate), issues detected by the satisfaction flow (how many problems were caught before they became negative reviews), time to resolve detected issues, new reviews per month per site, the most frequent topics in reviews (what happy buyers mention and what less satisfied buyers mention), and the dealership's reply rate to reviews received.
Read also how to get more Google reviews for your dealership.
The most valuable figure in the process isn't the average rating: it's the number of detected issues that were resolved before they turned into negative reviews. That number quantifies the real value of the up-front satisfaction flow.
Learning from the language of reviews
Google reviews, analysed in aggregate, are a source of operational information that many dealerships don't make use of. Buyers write about what matters to them spontaneously and without a filter.
AI can classify reviews by topic (speed of the process, clarity of the paperwork, condition of the car at handover, how the salesperson treated them, quality of after-sales) and generate a summary of the most frequent topics by site and by period.
Read also the KPIs a dealership must measure to control its business.
If three reviews from the last month mention delays with the paperwork, the problem isn't solved by replying more elegantly to those reviews: it's solved by reviewing the paperwork process and who is responsible for it. Reviews are the symptom; the process is the cause.
The end goal isn't to get more positive reviews more often: it's to deserve better reviews because the buyer experience has improved. Automation only speeds up a process that has to be genuinely good to deliver sustainable results.
More than 750 dealerships already use Dealcar for their daily operations
Dealcar records the handover date of each vehicle in the customer's file, which lets you trigger the satisfaction flow at the right moment from a real system event, not from a static list. Detected issues are opened directly in the CRM, linked to the buyer's file.
You can see how it works at dealcar.io/software-concesionario or request a demo at dealcar.io.
Frequently asked questions
Can you ask only buyers you know are satisfied for a review?
The recommended flow is to first ask every buyer how things went and send the review link only to those who reply positively. That isn't selecting who gets to give their opinion: it's resolving issues first and asking for the review at the moment the buyer's experience is genuinely good. What you must avoid is preventing a dissatisfied buyer from leaving their opinion: that isn't the aim of the process.
How many times can you follow up if the buyer doesn't leave the review?
One initial request and, at most, one reminder. Beyond that, pushing harder damages the relationship without improving results. A buyer who gets three or four messages asking for a review may end up leaving a negative review precisely because of the pressure.
Should AI reply to Google reviews directly?
It can prepare drafts for the person responsible to review and publish, but it shouldn't publish replies autonomously. Negative reviews, or ones that mention specific situations, always need human review before replying: a wrong reply to a negative review can amplify the problem instead of solving it.
