Pandora Advertising Costs in 2026: What $250 Buys and What Nobody Publishes

Two hundred and fifty dollars is the entire published price of advertising on Pandora in 2026. That is the minimum spend AudioGo, the self-serve platform SiriusXM Media runs, states on its own pricing and FAQ pages, and it is the only figure in this market that came from the seller rather than from somebody's estimate. Pandora publishes no CPM, no rate card, no daypart multiplier and no package tiers. An advertiser who wants to know what a thousand impressions will cost before the campaign runs is asking a question the seller has decided not to answer in public, and every guide that answers it anyway is filling the gap with arithmetic nobody can check.

Two Hundred and Fifty Dollars and a Card

The self-serve floor is $250 USD, or $300 Canadian, per campaign, stated plainly in the AudioGo FAQ. That figure doubles as the default billing threshold, so the platform charges the card when spend across running campaigns crosses it or at month end, whichever lands first. There is no insertion order, no signature and no media rep. A business with a finished audio file, a target radius and a card can be live on Pandora inventory the same afternoon, on a setup the platform advertises as five steps and under ten minutes.

What that floor buys in delivered impressions depends on the targeting attached to it, which is the part the published number conceals. A $250 campaign aimed at adults in one metropolitan area with no further filters clears at one price. The same $250 narrowed by genre, daypart and a predictive audience clears higher, because AudioGo states outright that behavioral targeting, Predictive Audiences and ad positioning each increase a campaign's CPM. The floor is fixed and what the floor delivers is not.

The Rate Card That Does Not Exist

AudioGo describes its pricing as an adaptive bid-based model that optimizes spend to secure relevant impressions at the lowest available cost. In practical terms the advertiser sets a budget and the system buys against it in an auction, which means the effective CPM is an outcome of the campaign rather than an input to it. A buyer learns the price after paying it.

Agencies buying Pandora at scale negotiate directly with SiriusXM Media and know their rates in advance. A Doylestown contractor spending $250 does not, and should treat any published Pandora CPM figure as an estimate with no source behind it. The planning method that survives contact with the invoice is to run the floor once with loose targeting, read the delivered CPM off the reporting dashboard, and size everything after it from that number. Our breakdown of what SiriusXM charges across its own channels runs into the same wall from the other direction, and the comparison with Spotify's self-serve audio platform is worth reading before committing a budget to either.

Forty-One Million, or Forty-Nine

Two pages owned by the same company give two different sizes for the audience being sold. SiriusXM Media's advertising page for Pandora states 41 million monthly listeners, of whom 87 percent listen with ads, generating more than a billion hours of listening a month across 202 US markets. AudioGo's Pandora page, selling access to the same inventory, states an ad-supported audience of over 49 million unique listeners. Both numbers are published by the seller. Neither page defines its measurement window or its counting method, and the eight-million gap between them is roughly the population of Virginia.

Neither figure is dishonest, since monthly listeners and unique listeners over a longer window are different measures. The gap does show the reporting standard an advertiser is working with. A buyer who would never accept an unsourced circulation figure from a newspaper accepts them routinely in digital audio, and the defensible position at this budget level is to buy on delivered impressions rather than on audience claims in a deck.

Four Products on One Invoice

SiriusXM Media sells four distinct things against Pandora inventory, and a budget split across them behaves nothing like a budget concentrated in one. Audio ads are the core product and the only one most small advertisers should buy. Display ads cover banners, web skins and rich media inside the apps. Video ads run at moments of active engagement, such as a login or a song skip, and the company claims full share of voice during them. Rewarded ads trade a stretch of free listening for a listener interaction.

The audio spot carries the structural advantage, because it plays whether or not anyone is looking at the screen, and a Pandora listener frequently is not. Display and video on an audio platform compete for attention that may not exist at the moment of delivery. On a $250 floor, splitting across formats buys a small sample of each and teaches less than one clean result in audio.

The Line Nobody Budgets For

AudioGo accepts audio up to thirty seconds in mp3, wav, aac, wma or ogg. That specification is the whole creative brief the platform gives, and it is where most of the money is lost. The platform offers a free synthetic voice tool that generates a spot from typed text, and a paid Creative Services option that turns one around inside two business days.

A synthetic read costs nothing and sounds like it. On a platform where the listener is wearing headphones, with the voice six inches from the eardrum and no competing noise, a text-to-speech delivery is identifiable inside two seconds, and the first thing it tells the listener is that the advertiser spent nothing. A $250 media buy behind a bad thirty seconds is $250 spent making people dislike a business, and the production that would have fixed it costs about what the media costs. That ratio is unusual in advertising and it is the most important thing to understand about buying audio at this level. Our guide to the craft rules behind a script that survives a car radio covers what a thirty-second script can hold, and what each part of a digital audio campaign costs breaks the two lines apart.

INSIDE VOICE, Thirty Seconds in Hatfield

Original :30 written for streaming rather than broadcast, fictional client.

ANNOUNCER (low, close, almost whispered): Don't react. You're at your desk, there are people around you, and I'm the only one who can hear this.
ANNOUNCER: You slept badly again. Same mattress, nine years, the dip on the left.
ANNOUNCER: Your back has been filing complaints since March and you keep telling it the mattress is fine.
ANNOUNCER: Keller's Mattress, Route 309 in Hatfield. Open till eight. Nobody has to know you went.
MAN (normal volume, from a few feet away): Who are you talking to?
ANNOUNCER (whisper): Nobody.

Seventy-two spoken words, and the premise only works on a streaming platform, which is the point of writing it. A broadcast spot is mixed loud and read to survive a car with the windows up. A Pandora spot usually plays into headphones on one person who is alone with it, so a script built for that fact can drop below conversational volume and address a single listener as a single listener. Most spots running on streaming inventory are broadcast spots the advertiser already had, mixed hot, shouting at somebody standing six inches away.

Eight Point Four Billion Against Everything Else

The IAB and PwC Internet Advertising Revenue Report for full-year 2025, published in April 2026, put US digital audio revenue at $8.4 billion, a 10.2 percent increase on $7.6 billion the year before, with podcast advertising inside that total reaching $2.862 billion on 17.6 percent growth. Those are healthy numbers in isolation and unflattering in context, because the same report put total internet advertising growth at 13.9 percent, video at 25.4 percent and social at 32.6 percent, leaving digital audio on 2.8 percent of all internet advertising revenue, the share it also held in 2022.

A category growing slower than the market it sits in, holding a flat share across three years, is a category where auction prices face less upward pressure than video or social. For a small advertiser that reads as opportunity rather than decline, because the listener on the other end is driving, working or exercising, and the money chasing video is not bidding for that moment. The case for streaming against traditional spots is laid out in our comparison of streaming and terrestrial radio.

Where That Leaves a First Campaign

Budget $250 for the media, because that is the published floor and there is no reason to exceed it on a first run. Budget roughly the same again for a produced thirty seconds, because the free alternative sounds free. Target loosely so the delivered CPM reflects something close to the open auction, read that number off the dashboard, and only then size a real campaign. Any Pandora cost estimate that did not come from your own dashboard, this article included, is somebody's guess about an auction they did not bid in.