Which Problems Go With One-Star Reviews at Car Dealerships?
Using 23,375 public reviews of 24 AutoNation dealerships, which sales and service problems go with one-star reviews, and why the store matters more than the brand.
A car dealer group reads its reputation as one star rating per rooftop, but each rooftop runs at least two businesses under one Google listing: the sales floor, with the finance office behind it, and the service lane. They fail in different ways, they are run by different managers, and a single average blends them.
To see what actually goes with a one-star review at a dealership, and how to compare rooftops fairly, we used public reviews of one large dealer group in its home market.
The data: 24 AutoNation dealerships in South Florida
23,375 public Google reviews of 24 AutoNation dealerships in South Florida, January 1, 2025 to June 16, 2026. The stores span 13 franchises, from Toyota and Chevrolet to Mercedes-Benz and BMW, grouped in the dataset into four tiers. 2,913 of the reviews, 12.5%, are written in Spanish, so they have to be read in the original rather than dropped.
The ratings are very high. 88.7% of reviews are five stars and the average is 4.62. Only 8.8% are one or two stars. Dealers ask happy customers for reviews, and often ask them to name their salesperson, so the five-star share says as much about review requests as about the visit. The useful signal is in the thin tail of bad reviews and where it collects.
Dimension Labs analysis
A high average with a small, concentrated bad tail
Public Google reviews of 24 AutoNation dealerships in South Florida, January 2025 to June 2026.
Which problems go with a one-star review
The analysis read each review for the kind of problem it raises and compared how often those reviews are one or two stars with the 4.5% rate for reviews that raise no problem.
Service breakdowns. A botched repair, a scheduler no one can reach, no loaner. 568 reviews, and 76.1% of them are one or two stars, about 17 times the normal rate.
Finance and fees. Add-ons, fees and the paperwork at signing. 1,327 reviews, 24.2% one or two stars, 5.4 times the normal rate.
Sales trust. Honesty, pressure and whether the deal stayed the deal. 6,396 reviews, 15.8% one or two stars, 3.5 times the normal rate.
From the AutoNation analysis
A service breakdown ends in a one- or two-star review 76% of the time
Share of reviews that are one or two stars, by the kind of problem the review raises, against 4.5% for reviews that raise none.
Rate and volume point at different places. A service breakdown ends in a bad review three times in four, but it is rare. Sales trust comes up in far more reviews, so even at a lower rate it accounts for more bad reviews in total: roughly 1,011, against about 432 for service breakdowns and 321 for finance and fees. A service director and a sales director looking at the same reviews should get different work lists: the service lane needs the phone answered and the loaner fleet sized; the sales floor needs the deal to stay the deal.
Dimension Labs analysis
The most likely problem to end badly is not the one behind the most bad reviews
The same three problems ranked two ways. Left: share of their reviews that are one or two stars. Right: estimated number of one- and two-star reviews that carry them (reviews times rate).
The same split shows up when reviews name a person. Salespeople are named in 4,209 reviews and service advisors in 3,720, and fewer than 4% of those reviews are unhappy. Reviews that name the finance manager (124), a sales manager (74) or the general manager (59) are unhappy 30.7%, 46.0% and 47.5% of the time. Customers name the front line when it goes well and the person above them when it goes wrong, which often points to a complaint that was escalated and not settled.
From the AutoNation analysis
Front-line staff are named in praise; managers and the finance desk in complaints
Reviews that name a staff member, by role: share that are unhappy (one or two stars), with how many reviews name the role and the share that are happy (five stars).
Reviews that name any staff member are unhappy 4.2% of the time, against 15.3% for reviews that name no one. Read that one with care. Part of the gap comes from how dealers ask for reviews: a customer asked to mention the salesperson by name is already a happy one. It is still worth tracking by store, because a rooftop where praise stops naming people is a rooftop where something has changed.
How to compare the dealerships
Across the 24 rooftops, the share of one- and two-star reviews runs from 4.1% at Volvo Cars North Miami to 23.6% at Mercedes-Benz of Pembroke Pines, which has 271 reviews. The obvious first cut is by brand tier, and mainstream stores do run worse overall: 10.0% against 8.4% for luxury and 5.7% for the Luxury Extended tier.
But the tier explains little. Toyota of Hollywood has 19.4% one- and two-star reviews on 1,659 reviews; Toyota Weston, the same badge in the same metro, has 5.3% on 2,578. Inside the luxury tier, rates run from 4.2% at BMW of Delray Beach to 23.6%. Weighted by review count, the tier accounts for about 8% of the variation between stores; the rest sits between stores in the same tier. With more than 1,600 reviews each, the Toyota gap is far too large to be chance.
Dimension Labs analysis
Stores in the same tier sit further apart than the tiers do
Share of reviews that are one or two stars. Dot: the tier's overall rate. Line: from its best store to its worst.
Dashed line: all 24 stores, 8.8%
So the fair comparison is store against store within a tier, and same badge against same badge where a group has more than one. Then read the worst stores by problem type: a rooftop high on service breakdowns needs a different fix from one high on sales trust. The same within-group logic applies to restaurants, where we grouped stores by the kind of location they sit in, in how to compare restaurant locations using guest reviews.
How a dealer group can run this on its own reviews
Split sales and service. Tag each review with the department the customer dealt with before looking at any rate. One listing holds both.
Tag every problem and every named role. Service breakdowns, finance and fees, and sales trust each get their own field, along with who was named: salesperson, advisor, finance manager, manager.
Use rate and count together. The rate shows how bad a problem is when it happens; rate times count shows how many bad reviews carry it.
Compare within tier and within badge. A luxury store against luxury stores, a Toyota store against the other Toyota store.
Read reviews in every language. In South Florida, one review in eight is in Spanish.
Discount the five-star share. Solicited reviews push it up. Watch the bad tail and the named-staff mix instead.
Product loop
The agent turns tagged reviews into the report and the deck
Screen recording of the Dimension Labs agent building a report and slide deck from analysed data.
Reviews come from customers who chose to post, and these figures show association, not proof that one problem causes a bad review. To size the money, join the tags to what the group already holds: repair orders and comebacks for the service lane, deal jackets and F&I product penetration for the sales side, and customer retention by store. The method for that step is in which complaints cause bad reviews.
FAQ
Why are car dealership ratings so high?
Dealers ask satisfied customers to post, often naming their rep. In this set 89% of reviews were five stars, so the average hides the problems. Look at the one- and two-star share and what those reviews describe.
Which department gets the worst reviews at a dealership?
Here, a service breakdown was the most likely to end in a bad review (76.1%), while sales-trust issues produced more bad reviews in total because they come up so often.
Does the brand decide a dealership's reviews?
Not in this data. Two Toyota stores in one metro sat at 19.4% and 5.3% one- and two-star reviews, and the tier explained only about 8% of the differences between stores.
Dimension Labs reads every review, survey, call and ticket for every rooftop, tags each one by department, problem and named role, and joins those tags to dealer data such as repair orders, deals and retention to show what drives revenue and repeat business. The agent then builds the store rankings, reports and alerts on a schedule. See automotive and multi-location intelligence, or start with one file of reviews.