🔧 Car Repair Quotes Are Overpriced 56% of the Time. Home Quotes, 37%.
Two Questions People Keep Mixing Up
How often does a trade overcharge, and how much does it overcharge when it does? Those are different questions with different answers, and almost every piece of consumer advice runs them together.
We can now answer both. Frequency comes from 3,394 analyzed quotes across 31 categories. Magnitude comes from 61 red-flag cases where we recorded both the quoted price and the fair market value. The two answers point in opposite directions.
Auto Repair Is Flagged More Than Any Other Category
55.9% of auto repair quotes are flagged as overpriced, 387 of 692. That is the highest rate of any of the 31 categories in our database, and auto is also the largest category, at more than double the count of roofing.
| Category | Quotes | Avg score | Flagged overpriced |
|---|---|---|---|
| Auto | 692 | 48.7 | 55.9% |
| Siding | 56 | 49.7 | 51.8% |
| Concrete | 88 | 54.5 | 47.7% |
| Plumbing | 243 | 53.4 | 47.3% |
| Roofing | 290 | 53.0 | 43.8% |
| HVAC | 205 | 61.0 | 31.2% |
| Electrical | 118 | 64.1 | 25.4% |
| Pool | 54 | 68.4 | 24.1% |
Roll the trades together and the gap is 18.8 percentage points: auto and collision work is flagged 55.5% of the time (N = 723), home services 37.1% (N = 2,671). Auto is not worst because it is biggest, either. Across the 31 categories, the number of quotes in a category does not predict its flag rate (Spearman rho = 0.13, N = 31).
But the Overcharge Itself Is the Same Size
Among quotes bad enough to earn a red flag, a mechanic does not take a bigger cut than a contractor.
| Group | Cases | Median score | Median multiple | Median overcharge |
|---|---|---|---|---|
| Auto | 33 | 20 / 100 | 2.75x | $1,600 |
| Home | 28 | 20 / 100 | 2.14x | $8,243 |
The multiple runs higher for auto but does not clear significance (Mann-Whitney p = 0.12), and the severity scores are identical (p = 0.99). What differs is the check. The median bad home quote takes five times more money, because the median home job in the archive is $15,000 and the median car job is $3,460.
The Dealership Premium Does Not Show Up
| Where | Cases | Median multiple | Median overcharge | 3x or worse |
|---|---|---|---|---|
| Dealership | 17 | 2.80x | $1,720 | 7 of 17 |
| Independent or chain | 16 | 2.45x | $1,525 | 5 of 16 |
Mann-Whitney p = 0.93. The worst quote in our auto archive is from a dealership: $9,000 for a catalytic converter worth $1,500. The second worst is from an independent shop: $4,500 for an A/C compressor worth $900. Same story, different sign on the door. With 17 cases against 16 we can only rule out a large effect, but a large effect is exactly what the folk wisdom predicts, and it is not there.
Why Frequency and Size Come Apart
Auto repair has the structural conditions for frequent overcharging. You are usually stranded, the diagnosis is invisible, flat-rate labor software produces an authoritative number in seconds, and almost nobody collects a second quote with the car already on the lift. Home services have accidental friction that protects you, because getting three bids on a roof is normal.
None of that changes the size of the markup. A padded quote is padded by roughly 2x to 3x in both worlds, because that is what a customer will accept before walking away. The ratio is set by human tolerance. The frequency is set by whether you had time to check.
What to Do With This
If you own a car, the base rate is against you: more than half the repair quotes we see are overpriced, and picking the dealership or the independent shop does not change what happens when one of them pads a bill. The only thing that reliably changes the outcome is knowing what the work costs before you say yes.
Methodology
Frequency figures come from the full QuoteScore database as of August 12, 2026: 3,394 quotes across 31 categories with five or more quotes each, published openly at /api/dataset/benchmarks.csv. Magnitude figures come from 61 unique red-flag cases published between April 25 and July 24, 2026, each with a recorded quote and an independently derived fair market value. The multiple is quote divided by fair value; group comparisons use Mann-Whitney U with tie correction. People submit quotes they already suspect, so absolute flag rates are higher than they would be in a random sample. The comparison between categories still holds, because every category is subject to the same selection.