Jared is Chief Strategy Officer at Consumable — this interview has been lightly edited for style and readability
I’m Jared Lapin, and I’m the chief strategy officer at Consumable. I help to find ways to use audio to reach new listeners in innovative ways, especially with Consumable’s Visual Audio Ad Unit, which is an interstitial in apps.
Consumable is an innovational audio company with a focus on finding ways to create signal-rich opportunities for audio advertising.
James Cridland: Podcast measurement is really having a moment at the moment. There’s a lot of talk from the Alliance of Measurement in Podcasting about plays and downloads right now. But I think you’ve got a slightly different view about measurement, haven’t you?
JL: I think it’s great that there are other people talking about accountability in podcasting. I think the things that they’ve been saying are along the right lines; they resonate with me for sure.
We’re focusing on understanding who that listener really is. Once the file is delivered, podcasting has some real limitations in terms of traditional digital identity. It doesn’t have a mobile ad ID, it’s disconnected in terms of play out, which is one of the things they’re talking about. So you don’t necessarily know who that listener is, even if that listener consumes the ad. So, we’re focused on that: how do you understand who that user is, and how do you make it so that you can maintain the relationship with that user after the download happened and after the (ad) playout happened?
JC: This is a thing called PODIQ, which Iyou are working on. Lots of “anonymized listener data from multiple touch points”. What does it actually allow the advertiser to do?
JL: What PODIQ does is it attaches the mobile ad ID to that podcast user - to that RSS feed user who typically only shows up with an IP address and user agent and that’s likely transient at that. We use our visibility into the programmatic ad space to understand who that user really is when they go to access that podcast. We create an identity for that user that’s persistent, so that you can do things like deliver visual audio ads to them after they’ve downloaded that podcast.
JC: Are you expecting podcast apps to support that? How are you actually identifying a user?
JL: We’re we’re doing it one of three ways. We have channel partners, podcast publishers, that run our prefix. So every time a user downloads a podcast from that publisher, we get the opportunity to see if we can match that user to an identity. We have a pixel so that an advertiser can do something similar. They can traffic a pixel with their ad campaign and build cohorts of the users who are exposed to their ads that are available in real time for all of the digital media strategies that you would execute with identity and marketing. And then, our third solution we call PODIQ Beyond. We see podcast avails that are presented programmatically across our marketplace. And we’re able to make that same match in real time across programmatic.
JC: What are you learning from those - particularly a prefix? You’re still just learning the same information as what a typical podcast download will learn. You’re learning a user agent and an IP address. What else are you actually learning from that then?
JL: We’re learning exactly like you said, we know the episode, we know the feed, we know the IP address and the user agent. But we also likely have some experience with that user, with that IP address and user agent in the relatively recent past, potentially the very recent past. So, in our PODIQ audience graph, we’re mapping that user to the identity that we saw in their last request, in their “fully featured” request, if you like - so that we can understand the other attributes of that user that weren’t presented in the podcast request.
JC: I guess you’re using some of the other databases which are out there - which tell you “are there children at this IP address”, and those sorts of things?
JL: That’s exactly right. But we would also know things like the apps that that user typically engages with. For advertisers who are using CRM data, we may be able to map it to their CRM so that that advertiser can speak specifically to that user instead of just generally.
JC: We talk a lot about identity, but actually this isn’t the user’s identity, is it? It’s information about that user, but still keeps them anonymous?
JL: It’s an anonymous identifier. But, it’s one that’s that’s persistent and it maps to things like ramp IDs, which are also pseudo-anonymized, but that then allow you to map them to all of these audience segments you’re talking about, and to really understand the breadth and depth of that user across their digital consumption.
JC: I might be listening to a self-help podcast about how to stop gambling, for example, but I don’t necessarily want people to know that I gamble. Or, you know, maybe I’m in Indonesia or I’m in the UAE, I’m listening to an LGBTQ podcast where um being LGBTQ is illegal. Um, what does it mean for privacy?
JL: We focus on the US with this solution. There are lots of privacy controls within the US and outside the US, and they’re all things that we respect as we enter markets. If you’re on your phone, this is happening already with your app exposures, with your mobile web exposures, with your streaming exposures. We always do it in a way which is compliant with the local laws. We certainly don’t want to create extra anxiety for a user: we’re not doing anything that’s unique in the digital ecosystem, just unique in podcasting.
JC: So we are talking about the mobile ID, which is on somebody’s phone, but Apple Podcasts doesn’t send that through. Most of the other podcast apps don’t send a mobile ID through.
JL: No, none of none of them really do.
JC: So I’m a bit confused as to how the mobile ID works with podcasting.
JL: So, Consumable sees that user in a lot of different environments, including their regular mobile app usage. When you’re playing Tetris or checking the weather in that app, if you’re passing your mobile ad ID, we would likely have that ad ID and know your IP and UA as well. And so we map it through that process.
JC: Okay. That makes that makes perfect sense.
JL: You’re absolutely right, though. The fact that RSS doesn’t really support mobile ad IDs or really any persistent identity is a challenge in podcasting in terms of sort of parity with modern digital marketing.
JC: So let’s assume that everybody in the world is using PODIQ to learn more about their listeners. What does that mean for advertisers? What can advertisers do that they can’t currently do right now in terms of podcasting?
JL: They can be more selective in the users that they speak to. So they can be more precise in their marketing efforts. The idea, my sort of fantasy of how advertising should work is that the messages should all be additive and there should be things that are relevant to me. And obviously having an identity and making that available to a marketer helps to do that.
You can do that obviously with identities, you can do that contextually, there are a whole bunch of ways, and it’s a question of which is most precise and which is most relevant for that for marketing effort. We really think that there’s a lot of opportunity in PODIQ in making that listener relationship more sticky for advertisers.
Everybody’s moving to more contextual advertising because they don’t love the identity solutions, as you pointed out. But podcast is very weak in every way. And so even a probabilistic identity that’s available in real time is really valuable because you can you can do something with it. Whereas right now, once that file is sent, you don’t really know very much about that person. And if they didn’t download it at home, you have no chance to know very much about that person.
Probabilistic identities are ones that are are assumed - there’s a whole bunch of of signals that triangulates on who that person is, whether it’s an attribute, an interest, or a person, right? Whereas deterministic identities, are where we know because you registered or you sent us your email address or you told us. The only way to be deterministic is to have them registered, ask them. And, there’s no way to do that with RSS.
If I insert a message in a podcast and it gets downloaded but never gets played, which I’m sure happens often - It’s not a particularly valuable execution. If I insert an ad in a podcast that that um does get played but is played to the wrong user, it may not be valuable. So having some visibility to know - did that ad performance actually happen? - is great. And with PODIQ at least you know who that user was that downloaded it. And so maybe you can speak to them in other environments as well.
JC: So you can catch the same person while they’re listening to a podcast, while they’re then using the New York Times app, while they’re then playing Tetris, as you say - with the same copy, with the same creative in all three of those of those environments as well, I suppose.
JL: That’s exactly right.
JC: Jared, thank you so much for your time. I really appreciate it.
JL: Thanks, James. I appreciate your taking the time to speak with me about it.