Formats look interchangeable on a rate card and behave nothing alike in practice. Five of ours are below, running. Press play on any one and you get the thing itself — the break, the skip, the pause, the scan — with the events we would report beside it.
Press play on the remote. The episode runs, the break arrives, and the ad takes the whole screen with no skip control anywhere — because there isn’t one. Pause the show instead and watch what fills the gap.
Living-room set · the same stream runs on every CTV device and OS
There is no skip button on an ad-supported stream, and the remote’s fast-forward is disabled for the duration of the pod. Completion sits a hair under total, which is why CTV is priced against reach rather than attention.
Stop the show and the paused frame makes room for a static unit beside it. It is the one CTV format the viewer asks for by pressing a button, and it runs for as long as they are away.
Breaks run two to four spots deep. First in pod reaches a room that has just stopped watching a story, last in pod reaches whoever stayed. Both are buyable, and both report separately.
Press play and watch the episode run. The ad is not baked into the file — it is stitched in at the break, chosen for whoever is listening, and measured from the first second to the last. Skip it and the measurement says so.
The episode carries markers, not ads. At each marker the server drops in whichever spot was bought for that listener, so the same episode serves a different advertiser to the next person.
The companion takes over the player and the lock screen for the length of the read, which is the only place a podcast listener can tap. It stays long enough to be acted on.
A listener can jump the break. When they do, the quartiles stop where they stopped and the skip is logged. You are buying measured listening, not assumed listening.
Radio runs live, so there is nothing to scrub and nothing to fast-forward. The break arrives, the spots play in order, and the only thing a listener can do about it is change the station. Pick where in the pod your spot sits and watch what that costs you.
In-dash head unit · the same stream runs on phones, desktops and smart speakers
Spots run back to back and the listener hears all of them. First in pod gets a fresh ear and carries a premium; last in pod reaches whoever has not reached for the dial yet. It is the one decision that moves radio performance most, and it is targetable.
The station's over-air break is replaced for digital listeners, which is what makes a broadcast station addressable at all. Same show, same break, a spot bought for the person actually listening.
There is no fast-forward on a live stream. Completion is near total, and the real risk is the dial — so tune-away during the break is the number worth watching, and it is reported.
Press start, turn your sound up and play. A sponsored message runs while you keep playing: the music ducks, a companion banner slides in, and every quartile is measured beside it.
Mute the device to see the audibility check: the ad is held until sound is on, and pauses if sound drops mid-ad.
The ad plays in the background. Game music ducks under the voiceover and comes back when it ends. No full-screen takeover, no pause, no skip button to hunt for.
A 320×50 companion slides in with the audio, so the ear gets the message and the eye gets a tappable destination for the rest of the session.
The SDK checks device volume before serving. Muted players get a prompt instead of a wasted impression, and quartiles only advance while the sound is on.
A reader presses listen instead of reading. A synthetic voice narrates the article and the ad runs before it. Pick a different article and watch the decision change — because here the words on the page are the signal, and sometimes the answer is no ad at all.
There is no listener history to lean on and none is needed. The page is read before the narration starts, and the spot is bought against what the piece is actually about.
The same scan that finds the match finds the reasons not to run. A breaking disaster story takes the slot off the market rather than putting a brand next to it.
The reader has the page open while they listen, so the audio carries the message and the in-article unit carries the click. It is the one audio format where both are on screen.
Fictional publications, shows, stations and advertiser · Voiceovers use your browser’s built-in speech · Measurement is simulated for illustration