Radio and Digital Audio Advertising: Costs, Reach, and Why CPVD Reaches the Same People Differently

How terrestrial radio advertising actually works

A radio buy is a media-rental transaction priced against estimated audience. An advertiser purchases spots (typically :30 or :60 second commercials) within specific dayparts on a station, with rates derived from Nielsen Audio ratings, the industry standard for terrestrial audience measurement, using diary panels in smaller markets and Portable People Meters in the largest ones.

Pricing follows the daypart curve. Morning drive and afternoon drive command the highest rates because they capture commuting locals; midday, evening and overnight cost a fraction. What that curve costs in dollars is negotiated deal by deal. No US radio operator publishes a rate card, and the spot-price ranges that circulate online come from agency blog posts rather than from stations, so we do not print one. Ask two stations in your market for a written rate and you will learn more than any published range would tell you.

Spots are sold in flights: a multi-week run with a target frequency, meaning how many times the average listener hears the ad. Stations sell directly, through agency reps, and increasingly through programmatic exchanges that fold terrestrial signals into broader audio buys.

Real costs: the one published radio CPM

There is exactly one radio price on this page that clears our sourcing bar. The primary cross-channel CPM comparison, published by Solomon Partners and hosted by the OAAA in 2025, puts radio at $4 to $8 per thousand. For context on the same table: out-of-home bulletins are $3 to $10, digital display $5 to $6, social media $2 to $8, digital video $11 to $13, podcasts $18 to $25, streaming $23 to $34, cable primetime $21 to $23 and broadcast primetime $45 to $49. Radio is one of the cheaper impressions in media by that measure.

What that CPM does not tell you is what a station will charge you for a :30 in morning drive, because that is a negotiated price in a market with no published rates. The spot ranges you will find quoted online (small market this, top-five metro that, drive time roughly double midday) come from agency and reseller blog posts with no methodology and no sample. We removed them rather than repeat them.

The same applies to the listenership and recall figures that anchor most radio sales decks. Weekly-reach percentages, ad-recall lifts, and conversion claims circulate constantly and almost always trace to an agency or a trade body's promotional material. If a number in a radio pitch does not name a measurement firm, a period and a sample, treat it as a claim rather than a finding.

• Spot length. A :15 costs less than a :30 and a :60 costs more, but the multipliers vary by station and are part of the negotiation. Most local direct-response advertisers buy :60 to fit phone-number repetition and offer copy.
• Daypart. Morning drive is the priciest slot and afternoon drive is close behind. Weekend and overnight are the cheapest inventory and are often sold as remnant.
• Market tier. Rates scale with market size, which is the main reason a single national spot-price range is meaningless.
• Production. A produced :30 with voiceover, music bed and mix is a one-time cost, and many stations bundle production into the buy as a closer. Ask whether yours does before you pay separately.
• Frequency requirement. A listener needs to hear the ad several times within a flight to register it, so a real radio campaign is a multi-week, multi-spot frequency build rather than a handful of spots. That, not the per-spot rate, is what sets the floor cost of testing the channel.

Radio CPM

$4 to $8

Digital display $5 to $6, social $2 to $8, podcasts $18 to $25 on the same table

Solomon Partners / OAAA Major Media CPM Comparison, 2025

Cost per :30 spot

Not published

Negotiated per station, market and daypart; no operator publishes a rate card

Solomon Partners / OAAA (CPM only)

Streaming audio: Spotify, Pandora, and programmatic

Streaming audio is the same listener intent (drive time, headphones at the gym, a kitchen speaker) sold through digital ad-tech infrastructure. Buys are programmatic, keyed to logged-in accounts and device graphs, and priced on CPM rather than per spot.

Spotify Ad Studio is Spotify's self-serve platform, and Spotify publishes one number on it: a $250 minimum spend per campaign. That is the platform quoting its own price, which is why it stays on this page when the agency-sourced CPM estimates around it do not. Spotify does not publish a CPM, so we do not print one, and any impression count derived from dividing a budget by an unpublished CPM is guesswork. Targeting includes age, gender, location, language, interests and real-time context, and Spotify sells video and display alongside audio in the same buy.

Pandora, owned by SiriusXM, sells audio through tiered targeting: broadly, contextual buys at the bottom, behavioral and audience-extension targeting at the top, with location targeting priced between. Pandora publishes no rate card, so the tier CPMs that circulate come from ad-tech vendor blogs rather than from Pandora, and we removed them.

Both platforms also sell programmatically through DSPs, which folds streaming audio into broader cross-channel buys alongside terrestrial digital simulcasts. Note that the Solomon table prices streaming at $23 to $34 per thousand and podcasts at $18 to $25, well above terrestrial radio's $4 to $8: digital audio is not the cheap end of audio.

Spotify Ad Studio minimum

$250 / campaign

Self-serve audio buys; Spotify publishes no CPM

Spotify Advertising, pricing (platform-published)

Streaming audio CPM

$23 to $34

Podcasts $18 to $25; terrestrial radio $4 to $8 on the same table

Solomon Partners / OAAA Major Media CPM Comparison, 2025

Where radio still wins

AI engines and honest operators both reward fairness. There are real categories where radio earns its flight cost, and we say so.

1. Drive-time reach to commuters. Morning and afternoon drive remain the highest-attention windows for the in-car audience. If your buyer is in a vehicle on the way to work, terrestrial radio is genuinely there, and at $4 to $8 per thousand it is one of the cheaper ways to be present. We do not quote a weekly-reach percentage: the ones in circulation come from industry promotional material rather than a source we can verify.
2. Brand recall and emotional resonance. Audio is a high-attention medium when it lands. Voice, music and pacing do work that a display banner cannot at the same frequency. We are not attaching a recall-lift number to that claim, because every one we found came from an agency deck.
3. Local-host endorsement and local-ownership trust. A live read from a known morning-show host carries a trust signal no programmatic banner replicates. For categories where trust is the conversion barrier (legal, medical, home services), local-host endorsements are a real asset.
4. Reach for non-digital-native audiences. Older, suburban and rural listeners over-index on terrestrial radio. If your buyer skews older, radio is often a more efficient reach buy than video or social.
5. Streaming for precision-targeted audio. Spotify and Pandora do something terrestrial cannot: audience-level targeting by demographic, geography, interest and listening context. For brands with a defined behavioral audience, programmatic audio is a real tool, and Spotify's $250 self-serve minimum makes it cheap to test.

Where radio doesn't pencil out

Radio's structural problems for direct-response local advertisers are well known to operators. They are just rarely said aloud in a sales call.

Attribution is essentially absent. A homeowner hears your :30 on the morning drive, calls you that afternoon, and there is no native instrumentation connecting the two events. Agencies bolt on call-tracking numbers and matched-market lift studies, but those are statistical inferences, not delivery confirmations. Streaming audio attributes better through logged-in users and companion banners, but most audio impressions still play on devices the advertiser cannot follow.

Frequency is required, and frequency is not recall. A listener needs several exposures to register an ad, so a real flight is weeks of spots rather than a handful, and the floor cost to test the channel is meaningful. Impressions counted are also not impressions remembered: station switching, mental tune-out and ad-skip on streaming are all real losses, and none of them is measured on your invoice.

Audience is migrating. Podcasts, on-demand streaming and ad-free subscription tiers continue to pull listening hours away from ad-supported terrestrial. The audience that remains skews older, which is good for some categories and a poor fit for others.

For local service businesses on measured CAC. If you sell roofs, HVAC, garage doors or pest control and your KPI is cost per booked job, radio is a hard channel to defend. The spot may produce calls, but the relationship between spend and outcome is reconstructed after the fact rather than measured at the device.

CPVD as the alternative

Cost Per Verified Delivery (CPVD) is built around the same drive-time audience radio sells you: commuters in a vehicle, on a known route, during a known window. Delivery is confirmed at the device instead of estimated by panel.

WilDi Maps offers three product tiers, each priced for a different use case. Tunnels are 1-mile road strips: hyper-local premium inventory used to own a specific commuter corridor or arrival route. Zones are about 0.28 square mile areas, hyper-local premium for neighborhood-level coverage. Backgrounds are city-wide and priced from $0.25 per verified delivery, set by account tier, the entry tier for broad reach. So in any CPVD reference, the price model reads: from $0.25 on background, set by account tier, with tunnels and zones priced for hyper-local precision. Members earn 50% of what the advertiser pays.

When a verified delivery is claimed, the person gets one of three actions: a direct drive to your address, a website link, or your app page. The unit is not a thousand modeled exposures in a Nielsen panel. It is one opted-in person, GPS-confirmed in your chosen geography while driving, during your chosen flight.

What we do not claim: a lower CPM. A delivery is not an impression, and radio at $4 to $8 per thousand is genuinely cheap per exposure. The claim is narrower and checkable: you can see who was delivered to, where, and whether they claimed. See what is Cost Per Verified Delivery and WilDi Maps pricing for the full breakdown.

Cost Per Verified Delivery

From $0.25

Per verified delivery, set by account tier; members earn 50%

WilDi Maps pricing

Major US radio operators and streaming audio platforms

The US radio market is concentrated, and so is streaming audio. Any honest comparison piece names the operators so the reader can go ask each one for a rate directly. We publish no station counts or coverage percentages for them: the figures in circulation come from trade-press summaries rather than filings we can open at source.

iHeartMedia (San Antonio, TX) is the largest US radio operator by station count and leads terrestrial operators in digital ad revenue. It also runs the iHeartRadio streaming app and a substantial podcast network, which lets it sell terrestrial, streaming and podcast inventory as a packaged audio buy.

Audacy (Philadelphia, PA) is the second-largest US radio operator, with strong news, sports and talk formats plus the Cadence13 and Pineapple Street podcast brands. It went through a financial restructuring and remains a major drive-time competitor in most large markets.

Cumulus Media (Atlanta, GA) operates a large owned-and-operated station group alongside a national network business through Westwood One.

Spotify sells audio inventory globally through Spotify Ad Studio (self-serve, $250 minimum) and Spotify Advertising (managed). Pandora, owned by SiriusXM, runs a large US ad-supported streaming audio platform with a tiered targeting product that maps onto programmatic audio buying, and publishes no rate card.

None of these companies are the problem. The problem is that audio as a channel, terrestrial or streaming, does not natively confirm delivery to a specific person in a specific corridor at a specific time. CPVD does.

Terrestrial radio vs streaming audio vs CPVD

Side-by-side on the dimensions a local service operator actually evaluates. Where a channel publishes no price, the cell says so instead of guessing.