Delivery verification
We check every placement after it runs: that it aired, that it ran as briefed, that it was disclosed and that it delivered what was bought. When it falls short, we claim the make-good.
Verification log
A row per placement with the aircheck or timestamp, each check passed or failed, delivery against guarantee, and any make-good claimed with its status. It updates weekly during a flight.
What we verify
-
The aircheck
For each podcast read we get the aircheck, the recording of the episode as it went out, and check it against the brief: length, position, talking points, offer, code and URL.
-
The integration
For each YouTube integration the log records the segment’s timestamp, length and position, the disclosure as spoken and as shown, the paid promotion label, and the description link.
-
Disclosure
Under the Endorsement Guides a disclosure must be hard to miss and must use the same means as the claim. A spoken claim needs an audible disclosure, a claim on screen a visible one, and a claim made both ways needs both (16 CFR § 255.0(f), retrieved 2026-10-05). A placement that fails is logged and raised with the seller the same day.
-
Delivery against the guarantee
Downloads or views are tracked against the guarantee across the agreed window. Agentio’s data puts close to 40% of YouTube integration views and 30% of clicks after day 30, and Agentio advises a 90-day assessment (Agentio, 2026-01-14).
-
How podcast delivery is counted
Podcast apps pass back little or no listening data. Delivery is therefore taken from server logs of files sent (IAB Tech Lab, 2024-05). We ask sellers for downloads counted to the current IAB Tech Lab guidelines, version 2.3 since September 2026 (IAB Tech Lab, v2.3, 2026-09).
-
The make-good
When a read runs short, drops the code, misses a required talking point or under-delivers, we document it with the aircheck or the view count and claim the make-good in writing.
Verification log
Flight 1, week 5
| Ref | Placement | Check | Delivered / guaranteed | Status |
|---|---|---|---|---|
| P-01 | Podcast B, ep. 212 | Aircheck: 61 s, mid-roll, code and URL read | 154,200 / 150,000 | Closed |
| P-02 | Podcast B, ep. 213 | Aircheck: 38 s, talking point 2 missing | 149,800 / 150,000 | Make-good claimed |
| Y-01 | Channel D, 06:10 to 07:41 | Disclosure spoken at 06:12 and label on | 268,000 / 450,000, day 30 | Open to day 90 |
| Y-02 | Channel E, 11:47 to 12:50 | Label on, no spoken disclosure | 201,000, day 30 | Raised with seller |
Why the length of a read is checked
Length is part of what was bought. Research by Nielsen for Spotify found the best host-read performance between 30 and 60 seconds (Spotify, 2021). In separate work, Nielsen measured roughly 40% more lift in purchase and recommendation intent from ads of 60 seconds and over than from shorter ones (Nielsen, 2021-09). A 60-second read that airs at 38 seconds is a different product, so it goes in the log and earns a make-good.
Terms used on this page
- Aircheck
- The published audio of a podcast episode, kept as evidence that a sponsored read ran to the brief.
- Make-good
- Additional delivery, or a credit, owed by the seller when a placement falls short of the agreed terms.
- Delivery window
- The span of time, fixed before purchase, in which downloads or views count toward a guarantee.
Start with one flight
Send us your category, budget range and what your finance team needs to see. We reply with how we would vet, buy and measure a first flight.
Submit a brief