Key takeaways
- Proof of play is a log from the media player saying it rendered your file — the device that plays the ad reporting on itself.
- Proof of display is independent confirmation that the ad actually appeared on the screen, from hardware separate from the player.
- A play log can be complete and correct while the screen was off, damaged, or showing something else entirely.
- The difference decides what you pay for: schedules and logs, or confirmed on-screen delivery.
Proof of play, defined
Proof of play is the DOOH industry's standard delivery evidence: the media player — the small computer driving the screen — writes a log entry each time it renders an ad file. Those logs are aggregated into playback reports: your ad "played" 1,240 times this week.
The limitation is structural, not dishonest: the device that plays the ad is the device reporting on it. The player knows it sent frames to an output. It doesn't know whether the screen was powered on, whether the panel was cracked, whether the input was switched to something else, or whether the screen was facing a wall in a closed room. Playback and display are different events, and a log only witnesses the first one.
Proof of display, defined
Proof of display raises the bar: a separate verification device — physically independent of the media player — observes the screen itself and confirms the ad actually appeared. The claim being verified isn't "the file was rendered" but "the content was on the glass". Because the observer isn't the player, a dead screen, a switched input or a failed panel produces a verification failure instead of a confident log entry.
Side by side
- Witness: proof of play — the player itself · proof of display — independent hardware watching the screen
- Detects a screen that's off: proof of play — no · proof of display — yes
- Detects a switched or faulty panel: proof of play — no · proof of display — yes
- Disputes come down to: proof of play — your word vs a log file · proof of display — an independent record
Why it decides what you pay for
Billing follows evidence. On a proof-of-play platform you're billed on the log — if the screen was dark, you find out (maybe) in a reconciliation argument after the money is spent. When billing is built on proof of display, an unconfirmed play simply isn't chargeable. On Admitt that's automatic: verification hardware is required on every listed screen, each play is checked against independent observation, and anything unverified is returned to your budget as credit without a dispute process.
How to test any platform in one sentence
Ask: "Is the device that verifies my ad the same device that plays it?" If yes, you're buying proof of play, whatever the brochure calls it. Independence of the witness is the entire difference.
Does proof of display cost extra?
It shouldn't. Verification that gates billing is only meaningful if it applies to every play — which is why on Admitt it's built into every campaign at no extra cost, rather than sold as a premium reporting add-on. If verification is optional, the billing isn't really built on it.