Interviewer: Hi everyone. Welcome to today's episode of The Sonic Truth. Today we'll be covering a hot topic in the audio space podcast measurement. We'll dive deep into 3 different myths around podcast measurement, provide some historical context around those myths, and offer a resolution that debunks these myths 1 by 1. I'm joined today by 2 experts in podcast ad tech and measurement. Allison Sprague from Wondery and Dan Jesson from ART19. Allison and Dan, welcome to The Sonic Truth.






Dan: Thanks so much for having us. Super excited to be here.






Allison: Absolutely. I'm a big fan of the show. Thanks for having us.






Interviewer: Of course. Before we dive into the content here, I'd love to learn more about each of your backgrounds. Do you mind giving us a brief introduction about you, your role, and your company?






Dan: Sure thing. I've been in podcasting for about 18 years now, and ironically I've never been a guest on a podcast, so I really appreciate the opportunity. I started my career in music as a bass player and recording engineer, but eventually pivoted to focus more on digital media and the surrounding tech. Before coming to ART19 I was at New York Public Radio, where I owned the tech stack for delivering all on demand and streaming audio as well as the monetization and data platforms surrounding it. It was a really awesome time to be in this space. It was brand new and the growth maturity I've seen happen on both the business and tech side over the past decade has been really amazing to witness and be a part of.






I've been with ART19 now for 6 years and we joined Amazon 2 and a half years ago. We're really focused on creating cutting edge tools for all podcast creators to help them grow their audiences, and maximize monetization opportunities. I'm currently head of product helping ART19 be the best home for podcast publishers, and a leader in the podcast tech space.






Interviewer: That's awesome. That's certainly quite the journey. Thank you so much for being here, Dan. We're glad to have you on. Allison, what about you?






Allison: I'm the head of ad measurement at Wondery, and Wondery Started out in 2018 with the goal of making Hollywood style podcasts, but quickly became known for its immersive storytelling style. In 2021 Wondery joined Amazon and has since become Amazon's global podcast content studio. I've been at Wondery for about 9 months, but I've spent around 12 years in the audio ads measurement space. Prior to working at Wondery, I led the audio measurement program at Amazon ads, and prior to that I started my career in radio at iHeartMedia. At the time, iHeart had just released their mobile app and they were looking for ways to bring digital style measurement to broadcast radio.






It was an exciting time. I got to build some of the first generations of ad measurement solutions for local radio, and to test and learn on some of the early podcast measurement solutions.






Interviewer: That's amazing. I bet you've seen a lot of change during your career in audio and podcast measurement. What are some of the most exciting advancements you're seeing today?






Allison: Yeah, I definitely have and I'd say I'm most excited about the advancements that podcasts have made in performance measurement. In the early days of podcasting, the measurement conversation was really around defining what should be counted as an impression, but now we're really focusing more on impact. What can a podcast do for an advertiser's business? And that evolution has been really amazing to watch and take part in.






Dan: Yeah, I completely agree with that. For the longest time, conversations about podcast measurement were really focused on defining, debating, standardizing, just how to measure a download and an impression. I distinctly remember being at an industry conference in early 2020 right before the pandemic, and there was a panel on measurement which for the first time was comprised solely of performance measurement companies. It was a refreshing change and a real turning point for the whole space. In the early days, there were lots of headlines around podcast downloads but being in the industry at the time, I knew there really were no standards for any of these claims.


That problem persisted for years, which I think helps spawn a lot of these podcast measurement myth that still exists today.






Interviewer: Absolutely, that makes sense. Thank you again for the background and context, so we often think of podcasts as a new medium and in many ways it seems like we're still at the beginning really of what podcasts can offer, but the conversation around podcast measurement and ad tech has been going on for almost 15 years now. This is a great introduction to today's topic, which is 3 myths of podcast measurement. Before we dive into the myths, Allison, what made you want to talk about this?






Allison: I really do have a passion for helping businesses attain measurable growth with advertising, and I wanted to talk about podcast measurement myths, because I hear these myths all the time from all different types of advertisers and agencies, and these myths play a role in how podcasts are bought and evaluated. Sometimes they may even hinder the ability for a brand to have measurable success and growth with podcasts, so I'm hoping today's discussion will help advertising customers feel more confident in using podcast ads to measurably grow their business.






Interviewer: Amazing. With that, let's go ahead and dive in with myth number 1. Podcast ads aren't as measurable as other digital media. Are they truly as measurable as other medias in your opinion?






Allison: This is a myth I hear almost every day from all different types of advertisers, and even from seasoned professionals in the industry, but it's time to bust this myth. Today, podcast ads are at parody with other digital media with regard to measurement.






Dan: Yeah, I agree but I think this myth maybe more than the others, really is rooted in historical truth. Five years ago the statement that podcast ads aren't as measurable as other digital media, was more accurate but you could have said that about OTT as well, so we've come a really long way.






Allison: That's exactly right. This myth is certainly well founded, but it's simply just not true today. To dive deeper, it might be helpful to look at 2 different types of measurement. Delivery and impact, delivery is the count of impressions, the audience demographics, reach, etcetera. The definitions of how podcast impressions are measured is different than the definition of how other digital media impressions are counted. This is part of what fuels this myth around podcasts not being measurable. However, the standards and the certifications that are required by server-side platforms for impression counting are at parody with display audio streaming TV, and other digital media. Impact measurement is more focused on ad effectiveness.






Or how did my podcast impact my business KPIs? Again, 5 years ago it would've been perfectly accurate to say that podcasts are not able to accurately measure business impact. However, today podcasts can measure nearly any KPI from awareness to web visits, online sales, retail traffic, incremental reach, TuneIn, you name it.






Interviewer: You talked about delivery and impact as 2 different types of measurement. Can you tell us more about how Wondery and ART19 measure each of these? Like what methodologies are used, what insights can advertisers obtain, etcetera?






Dan: Well, on the ART19 side, we're certified to measure downloads and impressions in accordance with the IAB version 2.1 guidelines, which are pretty much the prevailing industry standard for service side measurement today. What this means on the ad delivery side is we have a rigorous process to ensure ads are delivered to the listeners in full, and that is much invalid traffic as possible is filtered out before we count an impression. I think frequency measurement is also a relevant topic here and I'd like to touch on that for a minute as well, so frequency measurement is the ability to understand how many times the average listener was exposed to an ad. In podcasting we can provide two different types of frequency measurement. First is the average delivered frequency.






This is a common digital metric. It's derived from the total number of impressions delivered divided by the total household reach. Secondly, we can also provide advertisers with the frequency curve. Now this provides insight into the volume of listeners that received an ad at different frequency levels. For example with the frequency curve, advertisers may find that 10% of their impressions were delivered with 1X frequency and 25% of impressions were delivered at a 3X frequency, etcetera.






Interviewer: Oh, that's really interesting. I've heard of obviously frequency, but frequency curve is a new one for me.






Dan: Average delivered frequency is definitely the more common metric, but the frequency curve can also be helpful to understand campaign delivery post campaign. We can also provide frequency capping to help in the media planning phase. Frequency capping limits, how many times a user is exposed to your ad during a given time-frame. At ART19, we offer a first of its kind listener level frequency capping that's tailor made for high concurrency downloads. This means an advertiser can request a frequency cap of say 4 times per week, ensuring that the ad does not reach any single listener more than that 4 times in 7 days.






Interviewer: Frequency capping is possible in podcast advertising. Yet another reason why we're busting the myth. We talked about delivery measurement, but what about performance measurement? Is that also at parody with other digital media in your opinion?






Allison: Absolutely. Podcast advertisers today can measure the impact of their ads on nearly any business KPI, using the same methodology and solutions as they do with streaming audio display, streaming TV. Just about any other digital media. At Wondery, we have a suite of first and third party measurement solutions that we call Wondery ad analytics. We work with a variety of trusted third party measurement providers, including Veritonic, to offer a variety of solutions to our advertisers. Many of these solutions use the same deterministic and probabilistic methodologies for podcasts as they do with any other digital media.






Interviewer: Podcast ads can be measured with the same methodologies, and measurement solutions as other digital media. Let's now switch gears and talk briefly about optimization and how podcast measurements being used to make optimization decisions.






Allison: I like to say there's no point in measurement for measurement's sake. The purpose of measurement should always be to help inform future creative or media decisions. In podcast advertising, we have a variety of levers that we can measure and optimize toward. For example we may optimize towards different shows, genres, audience segments, creative tactics, ad placements, and more with the right impression volume, most measurement providers will allow advertisers to segment their results and derive actionable insights to inform optimization decisions.






Interviewer: Amazing, and what about cross channel measurement? Do these solutions allow advertisers to see how podcasts perform alongside other digital media?






Allison: Yes, absolutely. All of the solutions in the Wondery ad analytics portfolio offer cross channel media measurement. This allows advertisers to understand podcast performance in the context of their total media mix.






Interviewer: It sounds like we busted that first myth. Podcast ads are measurable with the same methodologies and solutions as other channels. Podcast ads can also be measured alongside other digital media for an understanding of total media performance, and the results are granular enough to be used for optimization and planning purposes, so let's move on to our next myth. The podcast landscape is always evolving as is the definition of a podcast itself. Myth number 2 is that not all podcast inventory is measurable, so before we dive into this myth, let's discuss first what are the different types of podcast inventory available today?






Allison: Great question. As you mentioned, the definition of a podcast is continuously evolving. We offer the following inventory types at Wondery. We offer dynamically inserted ads baked in or embedded ads, programmatic podcast inventory and video podcast inventory. We also offer custom content integrations and social extensions with our podcast shows and hosts, and I'll just come right out and bust this myth and say all of these podcast inventory types are measurable






Interviewer: Thanks for that, so let's take each of these 1 by 1. You mentioned DAI or dynamically inserted ads and baked in or embedded ads. What percentage of podcast ads are dynamic versus baked in these days?






Dan: Well, according to a recent IAB podcast revenue report, 84% or more of the podcast ads are dynamically inserted, so this means that baked in ads are a minority, but the volume is still there.






Interviewer: Okay, and for those that aren't sure, are all ads that are host red ads considered baked in?






xx: Actually no, host Red Ads can be dynamically inserted. This approach helps keeps the ad messaging current and relevant for the listeners, even if they are downloading a back catalog of episode.






Interviewer: Okay, so with dynamically inserted ads, those are measurable through pixels, right? What about baked in ads?






Allison: Yeah, dynamically inserted ads are measured either through a pixel or as we often do at Wondery, through direct integrations with the measurement providers and baked in ads, they're measured differently than DAI, but they're still measurable. There are a few different ways to measure baked in ads, but our preferred solution at Wondery is to build a direct data integration with a measurement provider. We find this method decreases risk for data loss and improves the customer experience.






Interviewer: Okay, so now we know both DAI and embedded ads are measurable. The next type of inventory you mentioned is programmatic podcast ads. Can you tell me more about how programmatic podcast ads work and how they're measured?






Dan: Sure thing, so while programmatic advertising is the dominant way, ads are bought and sold in other mediums, it's still early days in podcasting. That said, it's rapidly growing and according to the IAB, there's been 5X growth since 2021 with programmatic being tied to 11% of revenue industry-wide in 2023. Now for comparison in other media channels, that number is 87% so we've got a long way to go. The key advantages for programmatic is really ease of use for both the demand and the supply side. Publishers can control how their inventory is bundled and advertisers can buy across content all in a self-serve way. Today we support Wondery and other customers with direct SSP integrations through Vast. We also bring in Amazon demand to our targeted audience solution ad product through programmatic pipe.






Interviewer: Okay, and how are programmatic podcast ads measured?






Allison: Well, programmatic ads are inserted dynamically, so we measure programmatic inventory the same way we measure DAI through either pixels, or direct integrations with our vendors. We can measure just about any brand or performance related KPI on programmatic ads, so we'll use brand lift, reach lift, sales lift, traffic lift, you name it.






Interviewer: That makes sense, and the last inventory type you mentioned was video. Video podcasts have made a huge leap this past year. How do you measure video podcasts on YouTube?






Allison: Well, we think of video podcasts as any podcast that has a visual element to it, so you're right that a majority of video podcasts are consumed on YouTube, but we also think of using the format across social channels like TikTok and in-app video as well. Video podcast measurement capabilities are growing and evolving.






Interviewer: That makes sense, so how do you measure video podcasts?






Allison: Well, 1 option is to measure brand related metrics using a controlled or forced exposure methodology, and for conversion related measurement, 1 of the methodologies we are seeing is a modeled approach. While we aren't able to pixel a video ad or identify exposure, we can model video performance based off of audio performance. Of course, this is assuming that the audio and video podcast perform the same, which may or may not be the case, but it's one of the solutions that is most common in today's video podcast measurement landscape.






Interviewer: Interesting, so, so far we've busted 2 podcast measurement myths. Let's get into the third, which is a hot topic surrounding brand safety. There seems to be a belief in the market that brand safety controls are difficult to manage on podcasts. I'd love your thoughts on this.






Allison: This is another myth we hear a lot and it's a question we get frequently from advertising customers. At Wondery, we work with Barometer to measure brand safety and barometer's a third party, and they use the GM framework for brand safety and suitability with Barometer, we have GM compliant risk ratings for our entire catalog of Wonder Originals and partner shows.






Dan: That's right, and we've recently onboarded all of ART19's targeted audience solution content to Barometer as well. This provides GM scoring for the targeted audience solution product and allows us to provide media plans that are tailored to different brands risk sensitivities.






Interviewer: Got it. Now, in those answers, I heard you mention 2 different terms, brand safety and brand suitability. For those that may be less familiar, are these synonymous terms?






Allison: Actually no, they're different, so brand safety is the process of excluding content from a campaign. While brand suitability is the process of including content in a campaign. For example, a brand may find profanity not safe and therefore choose to exclude content with risk levels highlighted in profanity. This would be considered a brand safety decision. In contrast, a health food brand may want to include their message in content where the hosts are talking about healthy eating or healthy lifestyles. This would mean that the health food or healthy lifestyle content is brand suitable.






Interviewer: Thank you for that clarification, and podcast advertisers have access to both of these? Both brand safety and brand suitability measurement?






Allison: Yep. We provide advertisers with access to both brand safety and suitability measurement.






Interviewer: Why do then, do you think there's a myth about podcasting not being brand safe or brand suitable?






Dan: Well, this is another myth that is historically founded in truth. Five years ago there was no guidelines or frameworks for what constituted brand safe or brand suitable content. In 2019, the Global Alliance of Responsible Media, AKA GARM, was formed to help create these standards for all media types, and it wasn't until the past 2 years that GARM and the needed tooling has been available in podcasting to create a consistent, and reliable framework with which we can monitor rate and measure content against the GARM standards






Interviewer: Got it, and how many advertisers today would you say are using brand safety, or brand suitability measurement and how important do you think it is for them to focus on brand safety?






Allison: It's definitely an area of increasing interest for advertisers, so according to the 2023 IAB podcast revenue report, nearly 70% of podcast publishers report that they now use brand safety tools. This represents a 57% growth year over year, so it's definitely not a niche product offering, but to answer your last question, how important is it? I really think it depends on the campaign goals. If the advertiser has historically had issues with trust or favorability, then of course there may be a case for excluding shows or genres with certain risk levels and content categories, but for performance advertiser, we may choose to optimize toward conversion regardless of the safety and suitability of the ad surrounding content.






Interviewer: That makes sense. I would think some advertisers are a bit hesitant to advertise in certain content types, things like true crime or podcasts that talk about murder or violence. Do you think this is a valid concern?






Allison: Well, I think all brands have the right to choose what content is right or suitable for them and their audiences, but true crime is certainly an interesting topic for brand safety. A recent study from Sounds Profitable highlighted that nearly 80% of true crime listeners are comfortable with the content and true crime podcasts, and less than 10% of respondents said they didn't like hearing brand messages in true crime. At the end of the day listeners are opting into this content, and a diversity of content types help keep the podcast industry continuing to grow.






Interviewer: Very, very true. Brand safety and suitability is certainly a fascinating topic in an area I'm sure we're likely to see continued growth in in the coming years. Well, we certainly covered a lot of great information today and busted some key myths in podcast measurement. To recap, we talked about podcast's journey, to measurement parody, the difference between measuring baked in verse dynamic ads, and of course we talked about brand safety and suitability measurement in podcasting. We learned that these myths while founded in historical truths are no longer true for the podcast industry. Thank you again, Dan and Allison for sharing all of this great information with us today and for being my guest. It's been so wonderful.






Dan: Thanks so much for having us. It was a super fun time.






Allison: Yeah, thank you so much for inviting us to join you. Had a great time.






[END]