All reports
For Market Researcher For Dealer Group

The pre-earnings read on the public dealer groups, weeks ahead of the print

The pre-earnings read on the six public dealer groups plus CarMax and Carvana: units, a revenue proxy, price, turn and mix, weeks ahead of the print.

What you get

  • Scorecard: each group's new and used units, the new share, asking-price revenue (the summed advertised asking prices at exit, in total and by market, unknown rather than zero when the API sends no sum), the change in units over the window and the priced sample
  • Price and turn: the median asking price at exit in each market, its change over the window, and the median days to sell and days on the selling site, new and used
  • New and used vehicles by month, one line per group, with a month table per measure
  • Optionally one group's footprint by state for the month: units, new and used, the median asking price and days to sell per state
  • Pre-earnings read: the revenue-proxy leaders, the fastest and slowest on the selling site, and the biggest moves over the window
  • Key metrics: the revenue and volume leaders, the fastest and slowest on site, and the biggest unit rise and fall
See a sample report

You provide

Optional (2)
  • Months
  • State footprint for

Nothing required: the eight groups are fixed; optionally the months in the series (12 by default) and one group for a state footprint.

What your assistant shows back

The interactive report Claude and ChatGPT show in the chat, built on live MarketCheck data. Every run also returns a shareable link to the same report.

You

Pre-earnings read on the public dealer groups this month

Claude

via MarketCheck data

Try a prompt in Claude or ChatGPT

Each one opens a new chat with the prompt filled in. Swap in your own details.

Who runs the Public Dealer Group Scorecard

  • Equity and credit analysts covering the publicly traded auto retailers.
  • Lenders and investors tracking CarMax and Carvana against the franchise groups.
  • Dealership groups benchmarking themselves against the public names.

Before you run it

Where does the data come from?
The Sold Summary API at national grain: a sale is inferred when a listing leaves a dealer website, and each price is the advertised asking price at that moment, never a negotiated transaction price. Sales are inferred retail units from the groups' websites, not the same-store units the companies report; a directional pre-earnings signal, not a substitute for the print. The data runs through the latest complete month.
What is the revenue proxy?
The sum of the advertised asking prices at exit of the priced sales, in total and by market. It is not reported revenue: no negotiated discounts, no finance or service income, and the sales that carried no price are not in it. When the API sends no sum for a market the proxy is shown as unknown, never as zero.
Which groups are in it?
The six publicly traded US dealer groups (AutoNation, Lithia Motors, Penske Automotive Group, Group 1 Automotive, Asbury Automotive Group, Sonic Automotive) plus CarMax and Carvana, each resolved against the API's canonical list on every run; any name that does not resolve is named in the report. CarMax sells used only in this data.
What comes back?
For the latest month, each group's new and used units, new share, revenue proxy, median asking price in each market, days to sell and days on the selling site, and the change in units and in each median over the window (12 months by default); twelve-month lines per market; optionally one group's state footprint for the month; and three findings from the numbers.
What does it cost?
$24.99 per run. Connect the MarketCheck AI Desk to Claude or ChatGPT once and ask in plain English, or run it from this website. You pay per report from a credit balance; no subscription is needed.

Want programmatic access to the reports? Technical reference (inputs, outputs, the MCP tool name): Public Dealer Group Scorecard documentation.

Reports that pair with this one

Run the Public Dealer Group Scorecard

$24.99 per run. Run it here, or connect Claude or ChatGPT and ask in plain English.