A Czech podcast publisher that wants to show one figure for “total reach” across YouTube, Spotify and classic RSS runs into a problem. Each of those platforms measures something different, and the words “view” or “listen” do not mean the same thing on them. Adding them into one number means adding apples and oranges. The change that YouTube launches on 24 August 2026 deepens that gap, because public views there now start from the first frame of the video.
The following overview summarises what each platform actually counts, what a download means, what part of the industry agreed on around Spotify, and why metrics across platforms cannot simply be added together.
What counts as a view and a listen
The units differ in one thing: how much consumption they require before they count. At one end is a simple start, at the other tens of seconds of continuous listening.
| Platform and unit | What is counted | Threshold |
|---|---|---|
| YouTube views (from 24 August 2026) | video started playing | first frame, zero seconds |
| YouTube Engaged views | viewer stayed past the opening moments (distinguishes only Shorts) | undefined |
| YouTube Unique viewers | estimate of people, not plays | people, max 90 days |
| Spotify Play (from 11 June 2026) | episode watched or listened to for at least 30 seconds | 30 s |
| AMP Play | one real person listens or watches | 30 consecutive seconds |
| IAB download | file delivered to a device | roughly 60 s of data, listening not verified |
| Apple Podcasts Play | playback took place | more than zero seconds |
YouTube therefore sits at the loosest end after 24 August. Its public view says only that playback has started. The stricter figure that used to stand behind the public number moves into the Engaged views metric, and for long videos it equals the ordinary view count. The real size of the audience on YouTube is best described by Unique viewers, an estimate of people in which one person is counted once, even if they watch several times or on several devices.
What a download means and why IAB watches it
Classic RSS podcasts are still measured through downloads. The standard that determines what may be counted is published by IAB Tech Lab, the technical arm of the Interactive Advertising Bureau. Its Podcast Measurement Guidelines define a download as a unique download of a sufficient amount of data, roughly 60 seconds of listening, measured from server logs.
What matters is what a download does not prove. According to IAB, a download confirms only that a file was delivered to a device, not that the person actually played the episode. That is why the podcast advertising industry is trying to move measurement from downloads towards real consumption.
What the industry around Spotify agreed on
That shift received a formal framework last summer. The Alliance for Measurement in Podcasting (AMP), founded in July 2025, brings together twelve leading firms in the sector. In early June 2026 it ratified a definition of a podcast play as “one real person listening or watching for at least 30 consecutive seconds”. Dan Granger, CEO of Oxford Road, presented the alliance’s work in July 2026, for now as a draft open for comment.
Spotify aligned its definition with this threshold. From 11 June 2026 it counts a play as an episode watched or listened to “for at least thirty seconds”, and justifies this as “a reliable indicator of intentional consumption, filtering out accidental starts, and other low-intent activity”. The thirty-second threshold thus becomes the nearest common currency for real consumption, on which the platforms knowingly meet. YouTube, however, does not fall into it for long videos, because its only public number is now the first-frame view.
What a podcast actually is
The AMP alliance also came up with a definition of the medium itself, which a measurement offer needs in order to set its boundaries. It describes a podcast as “an on-demand show rooted in the spoken word that remains fully understandable as audio”. Radio Ink summed this up with a simple test: a podcast is what works with your eyes closed.
That sentence has a practical point. A video with spoken word that makes sense even without the picture falls under podcast measurement and can be compared across audio platforms. A video that loses its meaning without the picture is more of a classic YouTube show and belongs to other metrics. The boundary between a podcast and a video thus becomes a question of content, not of the platform on which the show runs.
Why the platforms cannot be added together
Metrics can be compared only within what they actually measure. Exposure figures, where a start or a file delivery is enough, include public YouTube views, Apple plays above zero seconds and classic downloads. They can be added up into a count of starts, but they can never be presented as audience or reach.
Actual consumption, which requires at least 30 seconds, is measured by Spotify Play and AMP Play. On YouTube the closest is the engaged view, which, however, works only for Shorts. An estimate of the number of people is offered by YouTube Unique viewers and by IAB’s audience metrics. These are comparable in intent, that is, “how many people”, but methodologically different.
The error arises when the public YouTube view is added to Spotify Play into a single “total reach”. One is a start, the other is thirty seconds of consumption, and the sum gives no meaningful number. It is more honest to report each platform in its own currency and to say that the numbers are not added together. For a brand buyer, that is a clearer basis than one big number built on mismatched units.
Comments