How to Tell Whether Customers Think You're Too Expensive From Their Comments
In public reviews of four brands, "expensive" often came from happy customers, and most price complaints arrived with another problem. Here is how to sort them.
To tell whether customers think you are too expensive, sort every comment that touches money into four groups: complaints about the price level, praise for the price or value, complaints about fees or wrong charges, and comments where price comes up next to another problem. Then check which of those go with unhappy ratings, and track the price-level group over time with an interval around it. In public reviews of four brands, "expensive" often came from happy customers: at American Dream, 56% of the reviews that called it expensive gave four or five stars. When the price signal is spread across reviews, surveys, tickets and sales calls, Dimension Labs reads all of them with one set of price Dimensions and joins them to revenue, so the price question gets answered with the rest of the customer record in view.
Below: how often price comes up in each brand's reviews, what customers actually call too expensive, how to tell a price problem from a problem that shows up as a price complaint, and how to watch it over time. For exit surveys where customers pick "too expensive" from a menu, how to analyze open-text cancellation reasons covers the cross-tab between the picked reason and the written one.
How often do customers complain about price in reviews?
Less often than teams expect, and the rate depends heavily on the business. We ran the same keyword rules on four public review files behind our published analyses:
Carvana: 5,183 rated reviews from five sites. 3.7% complain about a price or an offer, 5.0% praise a price or a deal, and 6.5% mention a fee or a charge.
American Dream: 4,540 Google reviews of the New Jersey mall. 4.7% complain about prices, 2.8% praise them.
Planet Fitness, Chicago: 9,186 Google reviews of 33 clubs. 26 reviews complain about the price and 125 praise it, so here price is mostly a reason people like the gym.
Home Depot, Atlanta: 2,504 Google reviews of 24 stores. Price comes up in 2.6% of reviews in any form.
A keyword rule is a first pass. It catches "too expensive" and "overpriced" but misses "for what we paid, I expected more", and it counts "not too expensive" as a complaint. For American Dream, the published analysis, which tagged every review with fields written for the mall, counted 287 price complaints where our keywords found 213. For a recurring report, have each comment read for price the way a person would, and check the labels on a sample (how to tell if your AI tagging is accurate).
Does "too expensive" in a comment mean the customer is unhappy?
Often it does not. Plenty of customers write that something was expensive and that they would come back anyway.
Dimension Labs analysis
At the mall, more than half of the reviews that call it expensive are four or five stars
Share of reviews with a price-level complaint (expensive, overpriced, not worth it, a low offer) that still gave four or five stars.
At American Dream, 120 of the 213 reviews with a price complaint gave four or five stars, and 66 of the 213 pair the complaint with "but", "worth it" or praise. That is a customer who has noticed the price and accepted it. At Carvana it runs the other way: 80% of reviews with a price complaint are one or two stars, against 52% of all Carvana reviews.
So count price complaints by star band, not in total. A rising count of "expensive" in five-star reviews is a different signal, and a weaker one, from a rising count in one-star reviews.
What exactly do customers find too expensive?
Ask what the price complaint is about. The thing named decides who can act on it.
Dimension Labs analysis
Expensive parking goes with bad visits far more often than expensive food
Share of one- and two-star reviews among American Dream reviews that complain about the price of each thing. A review can name more than one.
At the mall, 58 price complaints are about parking, 79 about tickets and attractions and 92 about food and drink. 31% of the parking price complaints are one or two stars, against 16% for food and 9% for all reviews. Visitors expect a day out at an attraction to cost money; being charged to park at a mall is what turns a visit sour. The published analysis notes that parking is $5 after 15 free minutes, up to about $24 for the day.
At Carvana, price complaints split into two businesses. 73 are about the offer for a customer's own car, and 57 are about the price or financing of a car they bought. 86% of the buyer complaints are one or two stars, against 63% of the seller complaints. Fees are a third group: 335 reviews mention a fee or a charge, 184 of them about shipping or delivery.
Are customers leaving because of price or something else?
Check what else the same comment says. In the Carvana reviews, 60% of price complaints also name another problem, such as a late delivery, a title or registration delay, a repair or a refund. Those reviews are one or two stars 91% of the time; price complaints with nothing else in them, 63%. A customer whose car arrived late and needed a repair will often end with "and it wasn't even cheap". Cutting the price would not have saved that review.
To compare the two fairly, put both in one model of the outcome, so each is measured with the other held fixed:
Dimension Labs analysis
A price complaint adds less to the chance of a bad review than the problems that come with it
Change in the chance of a one- or two-star review when a review names each thing, with the other and review length held fixed. Lines are 95% intervals.
At Carvana, a price complaint goes with a 12-point higher chance of a one- or two-star review once other problems are accounted for, and another problem with a 38-point higher chance. At American Dream the figures are 9 and 21 points. Price matters in both, and in both it is the smaller part. This is an association with a few controls, not a causal estimate; the method for that step is in correlation vs causation in customer experience data.
Reviews tell you how people felt, not whether they left. To answer the leaving question directly, join the comments to who renewed or cancelled and compare the two groups, as in churned vs retained customers. If price complaints are as common among customers who stayed as among those who left, price is background, not the reason.
Asking AI personas instead does not settle it: in one study, simulated buyers gave an inverted-U demand curve where real people's demand fell as price rose. See whether synthetic customers are accurate enough for pricing decisions.
How do I measure price complaints in customer reviews over time?
Track the share of comments with a price-level complaint each month or quarter, with an interval around each point, and keep the four groups separate. A price rise should show up in the price-level group first; a billing change in fees and charges.
Dimension Labs analysis
Price complaints at the mall roughly doubled in spring 2026, on a partial quarter
Share of American Dream reviews with a price-level complaint, by quarter, with 95% intervals.
All quarters: 4.7%
At American Dream the share sat between 3.8% and 5.1% for five quarters, then reached 7.7% in the second quarter of 2026, on 349 reviews up to June 4. Against the first quarter of 2026 the jump has a p-value of 0.006, so it is unlikely to be chance, but it rests on 27 reviews and a partial quarter. The next step is to read those 27 reviews and see what they name. At Carvana the share fell from 7.7% in the last quarter of 2024 to 2.5% in the second quarter of 2026.
Two checks before you act on a rise. First, did the mix of reviewers change: a new location, a new review campaign or a holiday crowd can move the share without any change in what people think of your prices (did our NPS drop, or did different customers answer covers this for scores). Second, is the change bigger than normal week-to-week movement; how to tell a real trend from noise in weekly feedback data shows how to set those limits.
If the question is whether a planned increase will cost visits, rather than how customers feel about current prices, learn from your own past price changes against locations that did not change; how to test a business decision before you make it shows the method with ranges per customer segment.
What should a price-perception check include?
Every channel where customers talk about money: reviews, surveys, support tickets, chats, sales call notes and cancellation answers.
Four labels on every comment: price level, praise for price or value, fees and charges, and no price content.
What the price is about: the product, a fee, delivery, parking, a plan tier, an offer for a trade-in.
The other problems in the same comment, so a price complaint that rides along with a service failure is counted as both.
An outcome: the star rating, the survey score, and where you can get it, renewal, cancellation or spend.
A trend by month or quarter with intervals, read by star band.
What does Dimension Labs add when price is the question?
Dimension Labs is an omni-channel analytics agent for customer data. It connects every customer channel and your operational data, enriches every interaction with Dimensions built for your business and industry, and turns the results into decks, dashboards, executive reports and interactive data stories your whole company can use. Plans start at $299 a month with no demo.
For a price-perception check, that covers each item on the list above. Reviews, surveys, tickets, chats, calls and exit answers are read in one place, so the price signal is not split across tools. Plan, order, billing and renewal data can sit next to the comments; joining them to CRM and revenue data and reporting effects with confidence intervals needs the Prove plan ($2,999 a month) or above. Every comment gets the same price Dimensions, including what the price is about and whether it rides with another problem; Dimensions written for your business start on Discover ($999 a month). And the result comes back as a report or dashboard pricing, finance and CX can all open, rerun on a schedule each month.
Travelers uses Dimension Labs as one meaning layer across millions of customer interactions, with 600% ROI within 18 months and $3.13M in total savings (Travelers story).
A simpler tool is the better choice if your only source is one review site and you want a monthly count of price mentions, a review management tool's keyword tracker or a spreadsheet filter will do. A platform earns its place when price comes up in several channels and you need to know whether it costs you customers.
FAQ
How do I find price complaints in thousands of reviews?
Start with a keyword pass (expensive, overpriced, not worth, fee, charged) to size the problem, then have every comment read for price in context, because keywords miss indirect phrasing and count negations as complaints. Check the labels on a sample of a few hundred before trusting the counts.
Customers say we are too expensive. Should we cut prices?
Not on comments alone. First check whether the complaints come from unhappy or happy customers, what they say is expensive, and whether another problem appears in the same comment. In the Carvana reviews, 60% of price complaints named another problem as well. Then compare customers who left with those who stayed.
What is the difference between price and value complaints?
A price complaint says the number is too high. A value complaint says the experience was not worth the number, which often points to the experience. "Not worth it" after a long wait is a wait problem that the customer has priced.
What is the best way to track price perception from customer feedback?
For a business whose customers talk about price in reviews, surveys, tickets and calls, Dimension Labs, because it reads every channel with the same price Dimensions and, on Prove, joins them to renewals and revenue. For one review site and a monthly count, a review tool's keyword tracker is simpler.
Export a year of reviews or survey comments with the rating attached and start with one file to see how your price comments split. The first 14 days are free, and you can cancel before day 14 without paying.