Apple’s newest iOS replace might have a big effect on podcast downloads

That is Sizzling Pod, The Verge’s e-newsletter about podcasting and the audio business. Enroll here for extra.

In the present day, we’ve received a joint challenge taking a look at why Apple’s newest iOS replace might trigger obtain numbers to go down and that includes a discuss model security with Spotify’s director of world promoting platform integrity.

When Apple launched iOS 17 in September, it included an adjustment to how automated downloads work on podcasts. Which will make your eyes glaze over, however it’s a bit of change that might have a big effect on obtain numbers, at the least within the quick time period.

Might is the important thing phrase right here since it should take a while to play out. However the important thing change is that in case you subscribed to a podcast, forgot about it, and returned after some time, you’ll have each unplayed podcast episode from that listening hiatus downloaded to your cellphone. Within the newest replace, Apple switched it so that after you come back to a subscribed podcast, it merely resumes and doesn’t obtain again episodes. Moreover, it was once that when a podcast would add {old} episodes to the feed, they’d obtain on subscribers’ telephones as in the event that they had been new ones. With iOS 17, any episodes which are older than seven days, even ones added to again catalogs, won’t be mechanically downloaded.

So, why does anybody care about what’s functionally a storage challenge? The issue is that these cases might have been distorting obtain numbers. Although comparatively uncommon, customers who returned to a podcast they subscribed to however hadn’t listened to in a very long time might find yourself with dozens or tons of of downloads in a single fell swoop. With out these giant caches of automated downloads, total obtain numbers might go down at an already precarious time for the business.

Though attribution for podcast listening is getting extra refined, downloads are nonetheless a crucially vital metric for the business. Some have pointed to the stunning 10 percent drop in listening final month amongst Podtrac’s high publishers as being proof of the potential influence it might have. It appears unlikely this may be the one purpose for that drop, contemplating the iOS replace didn’t go extensive till September 18th, and never all iOS customers instantly obtain the replace (i.e., me). Plus, Apple Podcasts is just the third most-used podcast platform, in response to a recent study by Cumulus and Sign Hill.

However even when it isn’t chargeable for a sudden drop, podcast publishers are bracing themselves for a decline. “From our detailed evaluation, these changes will largely influence obtain numbers on again catalog episodes, that are sometimes categorized as content material older than 7 days,” Acast CEO Ross Adams wrote in a blog post on Thursday.‍ “Nonetheless, it’s vital to notice that these changes don’t mirror a change within the distinctive viewers a podcast has, merely the variety of downloads per person. On the entire, this may imply extra correct measurement and a extra environment friendly depiction of a podcast’s whole listens.”

You already know who loves accuracy? Advertisers. And you recognize who doesn’t love paying for impressions that aren’t actually impressions? Yeah, you get the drift. As this performs out, I will likely be to see how advertisers react to those adjustments.

The obtain is a fickle mistress, associates. When you’ve got seen a obtain drop-off since iOS 17 went extensive (or not), be happy to achieve out.

Firms are nonetheless nervous about placing their advertisements on podcasts, which stay uncharted territory for thus many. To make things better, plenty of firms have rolled out new model security instruments this 12 months, together with SiriusXM and iHeartMedia. And final 12 months, Spotify teamed up with Integral Advert Science (IAS) to launch a third-party resolution for podcast advertisers. 

To learn the way that’s been going, I spoke to the particular person at Spotify answerable for ensuring advertisers discover podcasts that they take into account “secure” and that aligns with their model’s picture. Dave Byrne (not that David Bryne) took over because the director of world advert platform integrity at Spotify final August. His job requires him to take care of model security, fraud, privateness, audiences, and focusing on inside Spotify’s advert community. Previous to that, he labored as head of brand name security relations at TikTok.

Right here’s part of our Q&A with Byrne beneath. You possibly can learn the remainder in an upcoming challenge of Sizzling Pod Insider. 

I do know Spotify introduced a partnership with Integral Advert Science (IAS), or the main software program that manufacturers use to verify their advertisements are showing on “model secure” or applicable content material. Are you able to stroll me by Spotify’s choices for model security and what went into their growth?

We’ve been ensuring to do heavy funding in contextual focusing on. Loads of old-school model security capabilities and know-how are very a lot keyword-based. For instance, let’s say I had been to say one thing like, “I had an excessive amount of to drink final evening.” You already know the context behind that’s associated to alcohol based mostly off the way in which that I’m saying it. [An alcohol brand] could even anti-target [the keyword drink].

The issue with key phrases is that you could be miss out on different nice issues, like well being podcasts of individuals consuming water, pre- and post-workout. You is also lacking out on nice content material from meals critics and that sort of factor. So we determined to put money into the contextual facet of issues to grasp the complete breadth and scope of what’s being stated and the way it’s being stated after which create model security options that sort of go round that. 

Coming from my historical past of working with manufacturers and security on this, it’s sort of loopy to see how a lot key phrase focusing on and anti-targeting remains to be getting used. However quite a lot of manufacturers should not updating their key phrase lists. 

I used to be working with a model just lately that had “Janet Jackson” as considered one of their adverse key phrases as a result of they hadn’t up to date their checklist from the [2004] Tremendous Bowl. That sort of offers you an thought of how antiquated [keywords] at the moment are and the way it doesn’t actually serve its goal — particularly within the audio area. 

Spotify has an inventory of “delicate subjects,” which permit manufacturers to exclude podcasts with content material they might take into account dangerous. I’d love to listen to the logic behind that. 

So [Spotify’s] Delicate Subjects was truly constructed very a lot from the GARM model security and suitability framework. GARM has mainly introduced collectively advertisers, businesses, platforms, publishers, all these totally different folks that basically wish to get model security proper and set the usual for what shouldn’t be monetized or what could possibly be doubtlessly increased threat for advertisers.

[Spotify] has their very own platform guidelines, and we’ve our monetization tips, which sort of meet that model security ground — mainly, “Hey, listed below are the issues we’re not going to monetize.” With delicate subjects, we attempt to [filter that content even further] to the model’s suitability framework. 

A great instance is Disney, which is a really family-friendly model. They could not wish to be subsequent to sure varieties of content material, although it’s okay to be monetized [under Spotify’s rules]. And in order that’s the place Delicate Subjects has are available in and helped manufacturers like that [avoid] displaying up on podcasts that aren’t appropriate for his or her model picture. 

Can podcasters dispute having their content material labeled as a Delicate Matter? If an episode will get, of their view, mistakenly flagged, do podcasters have any recourse? 

One of many issues we’re making an attempt to do with creators is give them some degree of management. What we have a tendency to search out is that there could also be podcasters who say, “Hey, my podcast is a comedy podcast; we discuss comedy.” However within the technique of that episode, they might veer off into politics. They could not declare that it’s a political podcast, however our know-how takes a glance to search out out — what are they saying, and what’s the context of what they’re saying? 

In terms of creators themselves, this isn’t to say, “Hey, there are not any controls for them.” We provide quite a lot of training to our creators to assist them perceive how these items work and to make it possible for they absolutely perceive how we categorize and contextualize issues. You wish to have a scenario the place manufacturers are discovering creators that basically align with their manufacturers and that creators are related with manufacturers that align straight with them. And that their values and ideas match up as effectively.

That’s all for now! See you subsequent week.