If you run a dealership or a group, you’ve probably noticed that everyone’s rating looks about the same. Yours, the group down the road, the brand you compete with. 4.6, 4.7, maybe 4.8. It’s hard to tell who’s actually doing well.
This year’s research explains why, and it changes what “exceptional” means. The dealerships at the top of the market aren’t delighting more customers than good ones. They’re failing far fewer. And the number on your dashboard can barely see the difference.
The opportunity isn’t to collect more praise. Nearly everyone does. It’s to find out where your customers are being let down badly enough to say so in public, and to run the business on that.
What the data reveals: there is no middle, so your rating is a failure count in disguise.
Nearly nine in ten UK reviews are five stars. Seven in a hundred are one star. Everything in between adds up to under five percent. There is effectively no such thing as a moderately satisfied car customer: they either walk away impressed enough to advocate, or aggrieved enough to punish.
That should change how you read every average. Your rating isn’t a measure of how good your typical visit is. It’s a measure of how often you produce a disaster. And once you sort dealerships that way, the picture sharpens. Moving a site from good to exceptional raises its five-star share by about eight points, which is respectable and unremarkable. Over the same distance its negative reviews fall from about six in a hundred to under one, a seven-fold collapse. Across the whole market, from the best band of sites to the worst, the average rating travels six tenths of a star, a range most dashboards read as “about four and a half.” The share of negative reviews travels twenty-eight-fold.
Why it matters: the biggest groups aren’t the best ones, and the average hides why.
Scale doesn’t protect you. Among the 52 dealer groups with a thousand or more reviews, the bottom quarter by rating is the largest by footprint, with three times the sites of the top quarter. The best group averages 4.88, the weakest 4.06, and between them the one-star rate runs from under two in a hundred to one in five. One national group holds 4.77 across 187 sites, so size clearly isn’t the obstacle. Running a lot of sites doesn’t put a group in the bottom quarter; it only makes the choice more expensive.
Here’s the reporting consequence. A group reporting a national average of 4.64 looks like it’s performing at par. Given how spread out the market is, that same group is likely to contain a handful of sites below 4.0, and those are where the escalations and the churn live. Averages are the wrong unit for a business that’s bought locally. Negative feedback gives you a better one: which sites are collecting an unusual share of complaints, which are improving, which are moving the wrong way.
What changes: stop treating reputation as a report card. Start using it to decide where to focus.
The old model asks how the dealerships performed. The new one asks where to focus, what needs investigation, and whether the changes are helping. It works because feedback moves fast enough to act on. Between January and June the average rating slipped from 4.68 to 4.62, a drift most people wouldn’t notice. Over the same months the one-star share rose from 6.3% to 7.6%, roughly 550 more one-star customers a month, each leaving a permanent, searchable account. The rating registered a drift; the failure rate registered a fifth more disasters.
One more thing those failures do: they wait for an answer. The sector replies to 91% of five-star reviews and 84% of one-star, and in six months 3,605 negative reviews got no reply at all. Response rate won’t move your average (exceptional sites and poor sites reply at nearly the same rate), but an unanswered one-star review stands as the only public version of that story, and it’s the longest, most detailed thing written about your site. Reply because of who reads it, not because of what it scores.
Three ways the business runs differently
- Stop celebrating the average. Make the failures visible, site by site. Leaders see where the one-star reviews concentrate instead of assuming a 4.6 means every site is healthy. On Monday: a named list of sites, not a league table.
- Stop reading the group number. Read the direction of each site. Which are improving, which are moving the wrong way, and what changed. On Monday: this month’s movers, up and down, and a reason for each.
- Stop leaving the complaint as the last word. Answer every legitimate negative review. Not for the average; for the next customer who reads it. On Monday: the count of unanswered one-star reviews, by site, and a reply to each.
Where Reputation comes in
Know where to look. See the direction of travel. Catch what the average rounds away. Reputation shows the distribution of negative experiences across every site rather than the group average, tracks month-to-month movement, flags the negative reviews still unanswered, and surfaces the recurring problems behind the numbers so the conversation starts with where things went wrong, not whether the rating held.
The question for your next meeting: which three of our sites produce the most one-star reviews, and would our average have ever told us?
Three-sentence test. I learned that exceptional isn’t more praise, it’s far fewer failures, and the rating can barely show the difference. I’ll run the business on where the failures concentrate and which way each site is moving. Reputation could show me that across every site, continuously.




