Dealership Advertising Budget Allocation by Channel: What the Average Store Spends and Where the Mix Goes Wrong

Short answer: in 2025 the average franchised dealership spent 74.9% of its advertising budget on digital channels, led by search engine marketing (21.1%), third-party listing sites (20.0%), SEO and website (19.5%), and social (14.2%). Television took 10.5%, radio 6.9%, direct mail 5.6%, and newspaper 2.1%. Those are NADA's figures from dealer financial statements. They describe the industry; they do not tell you how to allocate your own budget. The right allocation for a store comes from the cost of an incremental sale by channel, adjusted for inventory and season, and most stores have never measured it.

The 2025 mix, dollar by dollar

NADA Data 2025 reports the average store's spend by medium as follows.

Source: NADA Data 2025, as reported by Dealership Guy, May 2026.

The trend is steady rather than dramatic. Digital was 72.2% of budgets in 2023, 73% in 2024, and 74.9% in 2025. Social is the fastest-growing line, up from 12.7% to 14.2% of budget in a year. Television has held around 10 to 11% for three years, which says something about how dealers feel about video reach even as the delivery shifts from broadcast to streaming. Note that NADA lumps connected TV into television, so the 10.5% understates how much of that money is now bought programmatically.

BIA Advisory Services, which forecasts local media buys only, sizes the 2026 digital side of dealer spending at $2.3 billion: $896 million in search, $548 million in video, $427 million in display, and $392 million in classified. BIA's total is far below NADA's because it excludes listing subscriptions, SEO retainers, and website costs, so treat the two as different lenses rather than competing numbers.

What each channel costs to buy a lead

Cost per lead is a poor final metric, for reasons we lay out in why cost per lead is the wrong dealership metric. It is still useful for comparing channels on the way in, as long as you remember that a lead from one channel is not worth the same as a lead from another.

The largest public sample is WordStream's annual benchmark across LocaliQ client accounts. For the "automotive, for sale" category, its 2026 report (13,474 Google search campaigns, April 2025 through March 2026) shows a median click-through rate of 8.28%, cost per click of $2.27, conversion rate of 6.01%, and cost per lead of $44.26. Cost per lead rose 13.9% year over year while conversion rate fell 22.6%, which is the pattern you would expect as AI Overviews absorb more informational clicks and the remaining clicks skew toward shoppers who already know what they want. Service and parts campaigns run cheaper: $29.96 per lead on a 15.5% conversion rate.

For Meta, WordStream's 2026 benchmark for automotive lead campaigns shows a $35.52 cost per lead on a 4.15% conversion rate, from a smaller sample.

Third-party listing sites do not publish lead costs. Dealer-reported figures collected by analytics practitioners put a healthy CarGurus account under $40 per lead and Cars.com under $50, with wide variance by market and inventory. These are practitioner benchmarks, not audited data.

The number that matters more is what a lead from each source is worth. Urban Science's 2025 registration-matched data shows internet leads close at 6% within 30 days, phone leads at 14%, and showroom leads at 25%, and that 16 to 20% of leads in each category buy from a competitor instead. A $44 search lead that closes at 6% is a $738 cost per sale from that channel before you count the sales it influenced without a form. A $36 social lead that closes at 3% is a $1,184 cost per sale. The cheaper lead is the more expensive sale.

How to allocate: cost per incremental sale, not cost per lead

The allocation method we use has four parts.

Split maintenance from growth. Maintenance channels keep the current pipeline full: brand search, listing sites, SEO and the website, retention email, and reviews. Cut them and volume falls within a month. Growth channels reach people who are not already shopping the store: connected TV, conquest social and display, direct mail into competitor ZIPs, streaming audio, and sales events. For most stores maintenance is 55 to 70% of the budget. The share goal decides the rest; the method is in how much should a dealership spend on advertising.

Price each growth channel on incremental sales. Not leads, not attributed sales, but sales that would not have happened without the media. For channels that can be tested by geography (CTV, mail, audio, social), a ZIP-level holdout is the cleanest method. For search, Google now runs incrementality tests at a $5,000 minimum, down from about $100,000, and Meta added incremental attribution to Ads Manager in 2025. A channel that cannot be tested should be funded at maintenance levels only.

Move money with inventory. Days' supply should drive the monthly mix more than the media calendar does. A store with 90 days of a slow model should push that model into listing sites, search, and social with real offers and pull growth spend toward it. A store with 25 days' supply of its best seller should stop advertising that model at all and shift to the models it needs to move. This is the single most common mistake we see: budgets set in January and left alone while the lot changes every week.

Move money with the calendar. Tax-refund season (February through April) is the most efficient used-vehicle period of the year; refunds averaged more than $3,700 in 2026 and pushed February used retail sales up 5.5% year over year. Model-year closeouts in August and September and the November to December stretch are the strongest new-vehicle periods. Budgets should lead those windows by three to four weeks.

An illustrative allocation

For a store spending $60,000 a month with a growth goal, an allocation built on those rules might look like this. The numbers are illustrative, not a recommendation for any specific store.

Two things about that table. Every line has a measurement that ends in a sale or a share point, not a click. And 47% of the budget is growth spending, which is high for a store at its natural share and about right for one chasing a share goal.

Where the mix usually goes wrong

Listing sites and search grow by default because they are easy to justify with lead reports, and they end up carrying 40 to 50% of the budget even when the store's own website converts better. Television survives on habit in some stores and is cut on habit in others, without anyone having tested whether it moves showroom traffic. Social is bought for reach and reported on engagement, which is a metric that dies at the checkout. And direct mail, the channel with the cleanest match-back to registrations, is dismissed as old when it is often the most measurable growth line a store has.

The fix is not a different average mix. It is a store-specific measurement plan, which we cover in how to measure dealership advertising ROI.

Questions dealers ask

What percentage of a dealership's ad budget should go to digital?

The 2025 industry average is 74.9% (NADA). Most franchised stores land between 65 and 85% depending on market density and brand. The share matters less than whether the traditional dollars, mainly TV, radio, and mail, are measured on sales.

How much should a dealership spend on Google Ads?

The average store spent $123,698 on search engine marketing in 2025, about $10,300 a month. A better guide is the store's own cost per sale from search, calculated by matching search conversions to the DMS, and its brand-search coverage. If brand and inventory searches are covered and cost per sale is under gross, spending more on search is rarely where the next unit comes from.

Is connected TV worth it for a car dealership?

For stores with a share goal and enough budget to test it, yes, provided it is measured against a holdout. In our own client work, a Volkswagen store measured an 80% relative lift in market share during a CTV program while cost per measured showroom visit fell. A neighboring luxury store on the same program gained traffic but not share, which pointed to a conversion problem, not a media problem. The channel works when the measurement is set up to catch both outcomes.

Should a dealership still advertise on the radio?

Radio was 6.9% of the average budget in 2025 and still reaches commuters in long-commute markets. The honest answer is to test it with a holdout or a promo-code match-back. If it cannot be measured to a sale in your market, fund it at a level you can afford to be wrong about.

Sources

• NADA Data 2025 via Dealership Guy; NADA Data 2024 via Inside Radio; NADA Data 2023 via Digital Dealer
• BIA Advisory Services, Automotive Tiers 1, 2 and 3 Local Ad Spend Insights, February 2026
• WordStream, Google Ads Benchmarks 2026 and Facebook Ads Benchmarks 2026
• Urban Science, Close Rate vs. Defection Rate, November 2025
• Digital Dealer, Why Automotive Advertisers Need Incrementality Testing, September 2026
• Cox Automotive, Used-Vehicle Inventory and Sales, February 2026