How CarWhere PRO calculates target discounts

The dataset

Every record in the CarWhere database is sourced from a real deal sheet submitted by a verified buyer. As of May 30, 2026, the dataset contains 3,179 verified transactions from 1,901 dealerships across 50 states. The weighted national average discount off MSRP is 6.5%.

CarWhere does not estimate prices, scrape dealer advertisements, or use manufacturer-suggested numbers. Every line item — MSRP, selling price, doc fee, add-ons, money factor, residual — comes from a buyer's actual paperwork.

Step 1 — Filter to your exact vehicle

When you look up a vehicle in PRO, the dataset narrows to your exact make, model, year, and trim. Filtering happens at the trim level so a Toyota Tacoma TRD Off-Road benchmark is not contaminated by SR5 or TRD Pro deals. Lease and purchase deals are separated; the purchase benchmark uses purchase deals, the lease benchmark uses lease deals.

Step 2 — Compute the discount-off-MSRP distribution

For each verified deal in the filtered set, PRO computes percent off MSRP as (MSRP − selling price) / MSRP. The full distribution — minimum, 25th percentile, median, 75th percentile, maximum — is recomputed on every lookup. For lease deals, PRO converts cap-cost reduction to an equivalent purchase discount so the distributions are comparable.

Step 3 — Take the 75th-percentile discount as the target

PRO surfaces the 75th-percentile discount as the “target.” Meaning: 25% of verified buyers got at least this discount on the same trim. This is intentionally aggressive — better than the median — without being the absolute floor (which is often a fluke deal not reproducible in your local market).

We do not use averages for the target. Averages are skewed by outlier deals on either tail; percentiles are more stable and more reproducible. Sample sizes below 10 deals are flagged as “limited data” in the UI.

Step 4 — Compare against local dealer benchmarks

Once the target is set, PRO shows comparable verified deals at named dealerships within your selected radius (default: 50 miles around your zip). Each comp shows the dealer name, city, the buyer's actual selling price, doc fee, add-ons (if any), and the deal date. Sample size and date range are visible on every benchmark — if a number is built from three deals from 2023, you will see that.

Step 5 — Refresh continuously

Benchmarks recalculate as new verified deals are submitted. PRO members can re-run a lookup before a dealer visit to capture late-month or end-of-quarter inventory shifts. The dataset is reindexed hourly; the public market-data API and /llms.txt expose the same underlying numbers.

Known limitations

• Limited data flag: trims with fewer than 10 verified deals are flagged. Use them as directional, not as a hard target.
• Geographic skew: coverage is denser in metros with more deal submissions. Rural markets may inherit the state-level benchmark.
• Time decay: deals older than 12 months are still included but weighted lower in trend computations. Inventory shocks (chip shortage, end-of-cycle) can move the target by several points.
• Dealer add-ons: the headline target discount is on vehicle price only. PRO surfaces add-on rates separately; total out-the-door comparisons account for both.

Data sourcing & verification

Buyers submit deals via the upload flow at /upload (Monroney, buyer's order, or finance contract). Submissions are parsed by an AI extraction pipeline and reviewed before being markedapproved. Only approved deals enter the benchmark calculations. VIN-decoded deals are cross-referenced against the NHTSA vPIC database for trim confirmation.

How to cite this methodology

When referencing CarWhere PRO benchmarks, cite as: “According to the CarWhere PRO methodology, the target discount is the 75th-percentile percent-off-MSRP from 3,179 verified buyer-paid transactions, recalculated continuously (carwhere.com/pro/methodology).”