Dealership advertising spend, 2012 to 2025: the total, the cost per new unit, and how the medium mix flipped

NADA has printed the same two charts in its annual report for longer than most dealership marketing departments have existed: total dealership advertising by year, and advertising per new unit sold. Read fourteen years of them together and the story is not the one the totals suggest. Franchised dealers spent a record $9.96 billion in 2025, but they spent it on fewer cars at higher prices, so the per-unit figure hit $739 while advertising as a share of sales fell below 1%. Meanwhile the medium table flipped: three search lines now take 60.6% of the average store's budget, and newspaper is 2.1%.

VulcanAX's own work is search and AI visibility, which is one of the four lines in that table and the only one that behaves like an asset. The series comes first, from NADA's 2019 and 2025 reports, then the mix, then what the per-unit number does and does not tell a store.

Dealership advertising spend in current numbers

Every figure below is NADA's except where marked as arithmetic on NADA's totals. The 2012 to 2019 series is from the 2019 report; the 2018 to 2025 series is from the 2025 report; the two overlap on 2018 and 2019 and agree. NADA's medium table changed categories between the two reports, from a single internet line to four digital lines, which is noted where it matters.

Fourteen years of the total, and the year the series broke

NADA's total for franchised-dealership advertising rose steadily from $7.52 billion in 2012 to $9.82 billion in 2016, the peak of the last sales cycle, then eased for three years to $9.25 billion in 2019. The pandemic year took it to $7.48 billion, a 19% cut in twelve months. The rebuild since has been uninterrupted: $8.19 billion, $8.57 billion, $8.90 billion, $9.22 billion and, in 2025, $9.96 billion, above the 2016 peak for the first time. The average store's share of it was $586,246 in 2025 against $554,292 in 2019, a 5.8% rise over six years that trails inflation.

The per-unit chart tells a different story over the same years, and it is the one a store should read. Advertising per new unit sold sat in a narrow band, $606 to $640, for eight years from 2012 to 2019. It fell to $582 in 2020, recovered to $602 in 2021 and then jumped to $718 in 2022, a 19% move in a single year, and has not returned: $708, $705 and now $739. The total did not jump in 2022; the unit count did. The average dealership sold 895 new vehicles in 2021 and 819 in 2022, the floor of the inventory shortage, while total advertising rose 4.6%. Fewer units, slightly more money, a permanently higher ratio.

More per unit, less per dollar of car

Set the per-unit figure against the price of the unit and the direction reverses. In 2018 the average franchised store sold 1,028 new vehicles at an average selling price of $35,608 and spent $624 of advertising on each, 1.75% of the price. In 2025 it sold 955 at $48,205 and spent $739, 1.53% of the price. The car got 35% more expensive; the advertising on it got 18% more expensive; the ratio fell. Against total dealership sales, NADA's 2019 report printed advertising at 0.9% of sales, and the 2025 report's own totals, $9.96 billion of advertising on $1.301 trillion of sales, put it at 0.77%.

That is the reading the total obscures. A record $9.96 billion is a record because prices and revenue are records; the industry is spending a smaller share of a bigger number. For a single store it means the per-unit figure is the benchmark to watch and the share of gross is the one to manage. A store whose per-unit advertising rose from $600 to $740 while its front-end gross per unit fell is in a different position from one whose gross rose alongside it, and NADA's average cannot tell the two apart. The budget benchmark on this site takes a store's own total and unit count and reports both readings against the published averages.

The mix flipped: 56% internet in 2019, four digital lines at 75% in 2025

NADA's medium table is where the six years between the two reports show most. In 2019 the average store put $310,127, or 56.0%, into a single internet line, with TV at 14.5%, radio 10.3%, direct mail 8.5%, newspaper 5.9% and other 4.9%. By 2025 NADA had split the internet line into four and the four took 74.8% of the budget: search engine marketing $123,698, third-party listing sites $117,249, SEO and website optimization $114,318 and social media advertising $83,247. Every traditional line lost dollars as well as share. TV fell from $80,428 to $61,556, radio from $57,037 to $40,451, direct mail from $46,893 to $32,830 and newspaper from $32,592 to $12,311, a 62% cut in six years.

Inside the digital three-quarters, the shape matters more than the size. Three of the four lines are search surfaces: the paid result, the marketplace listing and the store's own pages. Together they take 60.6% of the average budget, and two of the three are rented, in that the click stops the day the invoice does. The third, SEO and website optimization at $114,318, is the only line in the table that produces something the store still owns in year two, and the marketing post makes the case for why it is also the line least often asked to justify itself. The social line, $83,247, is the newest of the four and the one with the least automotive benchmark data behind it, which the Meta post covers.

The store is spending on fewer, more expensive sales

The unit series behind the per-unit chart is its own story. The average franchised dealership sold 1,028 new vehicles in 2018 and 1,026 in 2019, then 870 in 2020, 895 in 2021, 819 in 2022, 918 in 2023, 935 in 2024 and 955 in 2025. Seven years on, the average store is still 7% below its 2018 volume while its average selling price is 35% above. The dealer's advertising dollar is being spread across fewer transactions, each carrying more revenue and, for most of the period, more gross, which is why the per-unit figure could rise 18% while the share-of-sales figure fell.

The cost pressure on the number in 2026 is visible in the dealer surveys. Cox Automotive's third-quarter sentiment index put the costs sub-index at 71, the highest of the sub-indexes it reports, with 33% of dealers naming expenses as a factor holding business back, and one GMC dealer's verbatim listing advertising costs among the lines going up. The affordability post covers the customer side of the same squeeze. The advertising budget is being asked to produce the same units from a buyer pool that is older, richer and smaller than it was, and the per-unit figure is where that shows first.

The automakers above the dealer are cutting, and the industry around it is growing

Two forces frame the dealer's line. Above it, the manufacturers are pulling back: iSpot measured automotive national linear TV spending down 16.7% in the first half of 2026, to $978.8 million from $1.2 billion, and forecast the category falling below 10% of total U.S. ad spending for the first time after a 7% contraction in 2025. Around it, the market is growing faster than it has in years: the IAB raised its 2026 forecast to 12.3% growth in September, with social up 16.5%, connected TV up 15.6%, paid search up 8.1% and linear TV down 1.5%.

The dealer's 2025 mix already looks like the IAB's 2026 forecast: the growing lines are the ones NADA shows taking share, and the shrinking one is the TV line that fell from 14.5% to 10.5%. What the manufacturer cut changes is the air cover. A store whose brand demand was partly bought for it by national television is now buying more of that demand itself, on the same $739, which is one reason the per-unit figure has not come down as volumes recovered. The trends post sets the nine shifts behind that side by side.

How to use the number without being used by it

Three readings, in order. First, the store's own advertising per new unit over at least three years, against its front-end gross per unit over the same years; the national figure is the reference and the store's direction is the finding. Second, the share of sales, which for the average store is now under 0.8% and which a store above 1.5% should be able to explain by a market entry, a franchise change or a deliberate growth push. Third, the mix against NADA's, not to copy it but to know what the store is buying: a store at the average with 75% on search surfaces owns a page and a presence at the end of the year, and a store at the average with 40% on broadcast owns a memory. The benchmark tool runs the first and third readings from two inputs.

What the number will never do is say whether the advertising worked, because NADA counts spend and not outcomes, and the outcome that matters, a unit sold at a gross the store kept, is on the store's own statement. The published averages describe what dealerships do. They are useful because they are stable and public, and they are dangerous when they become a target, because $739 spent on rented demand and $739 spent on owned demand are the same line in the table and different assets in the store.

Where VulcanAX fits

VulcanAX's own work is the one line in NADA's table that leaves an asset behind, SEO and website optimization, and the answer layer above it: whether the store is named when a shopper asks a search engine or an AI assistant, on pages the store owns. The baseline audit reports where the store stands on that, and the budget benchmark puts the store's own figures against the averages on this page.

The paid search, listing, social and broadcast lines stay with the store and whoever runs them, and a store that wants those handled alongside the search work can raise that through VulcanAX. The series on this page is here because the industry's most-quoted advertising number, $739 a unit, means something different from what it is usually taken to mean.

FAQ