A Deep Dive into Programmatic – Part II (KPIs)

At first glance, Programmatic advertising may seem as complicated as trying to solve a Rubik’s Cube, understanding quantum mechanics, or even how Google works. But don’t fear, Granular is here!

We are departing on a five-part series of deep dives into the inner workings and magic of Programmatic. During Part II, Anna and Chris closely examine Key Performance Indicators (KPIs) and discuss the specifics of how they differ based on the channel.

Welcome to Getting Granular

The podcast where digital marketing experts from the agency Granular talk about the latest trends, tried and true best practices, and share their unfiltered thoughts about the digital marketing industry.

Programmatic Digital Strategy

https://player.simplecast.com/2a0c9b06-bbfd-4ae1-8d64-7069f85ae042?dark=false

SUMMARY

At first glance, Programmatic advertising may seem as complicated as trying to solve a Rubik’s Cube, understanding quantum mechanics, or even how Google works. But don’t fear, Granular is here! We are departing on a five-part series of deep dives into the inner workings and magic of Programmatic. During Part II, Anna and Chris closely examine Key Performance Indicators (KPIs) and discuss the specifics of how they differ based on the channel.

SHOW NOTES

  • A high-level look at what is a KPI (01:17)
  • The KPIs of OTT & CTV (03:39)
  • The main KPIs of audio (09:41)
  • The main KPIs of display & native (12:36)
  • The main takeaway (13:54)

New to Programmatic? Check out our primer episode Email us at info@GranularMarketing.com Follow us on Facebook

EPISODE TRANSCRIPTION

Narrator: Welcome to Getting Granular, the podcast where digital marketing experts from the agency Granular talk about the latest trends, tried and true best practices, and share their unfiltered thoughts about the industry. Whether you are here to learn how to grow your business, improve your digital skills, or just want to hear some Midwest PPC experts rant about digital media, you’ve come to the right place.

Chris: Thank you for tuning into the getting Granular podcast. I am your host, Chris Caesar, senior manager of paid media here at Granular. And we are joined by Anna Borchard again today as we continue our programmatic advertising podcast series, where today we’ll be covering programmatic KPIs or key performance indicators, what they are, why they’re different than what we would expect from a Google search or a Facebook campaign, and things that we can do to optimize for best performance in these specific ad platforms. Welcome in, Anna.

Anna: Happy to be back, Chris. Obviously we know KPIs across all advertising platforms and within the industry are important. So I’m excited to talk about what makes programmatic different.

Chris: Cool. So before we dive too far in, let’s just take a quick high level look and explain what the general KPIs are.

Anna: Well, it’s really going to depend on which channel you’re advertising on and especially what devices might come from some of those. So CTV is going to have some different KPIs than just a video or a OTT campaign, just based on the device, because you obviously can’t click on a CTV when you’re watching something on your Roku. An audio campaign obviously is also going to have different KPIs from video. And then we’ve got our more traditional display or native campaigns that again are going to have more of our standard KPIs, but those are going to be different from the out of home, the CTV, or the audio.

Chris: So let’s go over one example use case for a different type of campaign or scenario. So, for example, client A is upset because we’re running audio campaigns, but they’re not seeing any clicks on those ads. How would you explain that?

Anna: Yeah, so for audio it depends on the device, so we will potentially see some clicks depending on, again, if someone’s on their phone.

Chris: So for example, say I get a Pandora ad. Is that something that would come up? Spotify?

Anna: Yeah, it just depends on are we on a smart speaker? Are we on a desktop? Are on a tablet or phone? Because that is going to make a difference. Because when you’re on your phone, obviously you can pass through the ads on Spotify or pass through the ads on Pandora, unless it’s a non-skippable ad. Whereas then, when you’re on a smart speaker, you can’t necessarily click on anything. So in that instance, that’s where another KPI comes in.

Chris: Which would be?

Anna: It would be a completion rate. So completion rate goes across video. And then also some folks might be familiar with video completion rate or just videos completed on other channels. Obviously YouTube, Facebook video has those as well. So it would really just be that percentage of the total ad that was listened to in this instance.

Chris: Okay. So let’s take that a step further and just start to break down the individual channels and what the main KPIs are going to be for each one. So let’s start off with video, between OTT and CTV. What would you say are the main KPIs and give us a quick rundown of each one of them?

Anna: Yeah. So as I mentioned earlier, in some of the general KPIs, there will be a difference between OTT and CTV, just because depending on that device that you’re on, you either can click on it or can’t click on it. So, of course, if I’m watching the challenge on my Chrome cast, I can’t bypass that commercial, just because there’s nothing for me to click except to exit it. So on CTV, the main ones are going to be videos started, videos completed, and then the percentage of those two. So that video completion rate. So normally on CTV, we’re going to see about a 97% video completion rate.

Anna: Whereas on OTT, obviously you can have different devices that you can actually click on. So you can click through to the website, you get that URL or landing page to add where you can add UTM parameters. So on there we’ll see a little bit more clicks, not necessarily a lot. So the click-through rate will be pretty low. But again, that video completion rate is going to be that main thing. On OTT it’s more around 75 to 80% in that benchmark that I’ve seen across industry.

Chris: So why 97 and not 100?

Anna: Because you can exit on CTV. I definitely am guilty of it where I’m watching something on YouTube TV or on Hulu, or back when I had Sling TV, and I’m watching something and I decide that I just don’t want to watch it anymore. And then I leave in the middle of that commercial.

Chris: Sure. Okay. So then the lower video completion rate on OTT would be explained by not only those people who may just give up watching whatever they’re watching, but also they may have clicked on the ad and followed through to the website and potentially purchased or converted.

Anna: Yes, exactly.

Chris: Hopefully.

Anna: Hopefully they did. Yeah. So those are the main ones. A couple other things that we can look at for KPIs, I would call these potentially softer KPIs depending on what your client is looking for or what you’re tracking. But obviously frequency, we can keep that high or low. I am sure everyone has seen multiple ads a kajillion times. So that can either be a good thing or a bad thing, depending on what the purpose of the campaign is. If it’s awareness or getting low CPMs, then I would say it’s a good KPI. But of course, if you’re really just wanting to hit one-to-one, then it would not be the KPI you want to look at.

Chris: What would you typically recommend as a good frequency? I know that this is something that I have discussions with clients on all the time of sometimes, like you said, if we just want to reach somebody once, ad recall is going to be generally very low on that. So what would you throw in as a goal frequency to be hitting with people?

Anna: It depends on the budget, then also the client. I know that’s a non-answer answer, but there are campaigns that I set up where I’ve set up an unlimited frequency, because a lot of what CTV is based off of is being at the right place at the right time and having the right bid. So you can potentially not even hit that “unlimited frequency” if you’re putting it in. So I generally will bid for two per hour because you don’t always necessarily get that then.

Chris: Sure. Yeah. And, personal example, I’ve had many times where I’ll be streaming a sports event, a basketball game, whatever it is, and it’s just the same commercial every commercial break because, and you can correct me if I’m wrong here, it’s a combination of those people are going for an unlimited or a very high frequency, but also could it be possible that they’re just one of the only people filling in those available ad slots?

Anna: Yes, exactly. So that next KPI or softer KPI that we were going to talk about was the placement. So that would be either, of course, the website or the deal that is being placed for that. So for example, you can place ad specifically on March Madness, or the NFL draft was an example that I saw recently. So if there’s only a certain number of people selecting those particular placements or deals, then in theory, yes, that could be why.

Chris: Okay. Any other secondary or tertiary KPIs that you would also look to optimize around?

Anna: Yeah. So the other two would just be CPM. So your cost per milli. So obviously they’re going to be a little bit higher on CTV. I think we talked about that in our OTT/CTV podcast, but we’re generally just going to see higher ones. So if you’re not trying to spend $90 as a CPM, then that’s something that you want to keep an eye on. And then the same thing is your cost per view. You don’t necessarily want it to be too high if you’re trying to stay within a certain budget.

Chris: So a lot of my experience with cost per view reporting has come from YouTube where we’ll see cost per view is 2 cents, 5 cents, 7 cents, that really super low range. So would that be on par with what you would expect from a CTV or an OTT campaign or would you expect to see some variance there?

Anna: Yeah. I actually have a great use case for this. We’re running a campaign for a client right now where we’re running Facebook, YouTube, and CTV campaigns. So we’re seeing pretty much right around the same cost per view, which is two to three cents. So it can be pretty in line. Of course, a lot of that is going to depend on what your overall CPM is. We’re not just trying to target Hulu. So of course, if we’re just looking at some of those more premium placements, we’re potentially going to have a higher CPV just because we one, might not get as many views, and two, just have a higher CPM.

Chris: Okay, awesome. Let’s move on to audio. So this would be your, your Pandoras your Spotifys…

Anna: Podcasts, anything audio related. Yeah.

Chris: Awesome.

Anna: So similar to video, and we obviously had mentioned a use case for this, but depending on the device that the person is on, completion rate. So audio completion rate or just completion rate, you can call it, that would be the percentage of the audio started versus the audio completed. So generally 75% is what I’ve seen on those. And again, it can depend on what device you’re on. When I’m at home, I pretty much only listen to my smart squeaker and I just have it on in the background and don’t really pay attention to it. Whereas, of course, if someone’s running or in their car or doing something on their phone or watching or listening to Pandora on their desktop, they might have the opportunity to click on that ad or just bypass it like we did talk about with the OTT and CTV.

Chris: Okay. So this may be a better question for our podcast advertising podcast coming up. That’s that is very weird to say podcast advertising podcast, but anyway. A specific example that comes to mind for me is, so if I’m listening on apple podcasts and then I get what I would call a pre-roll commercial before the podcast starts, and then there’s a mid roll one, and then the same type of thing at the end. If I skip over those, A is the advertiser paying for that? B how does that get reported? Obviously it’s an audio started for lack of a better term.

Anna: Yeah. It would be an audio started.

Chris: An audio started. And I’m continuing to listen after that ad plays, so would it be counted as a completion or would it somehow know that I skipped over it?

Anna: It would theoretically know that you skipped over it, because there’s a difference between host-read obviously, where it’s live-read, and then more of the dynamic as we’re talking about with the pre-roll or the mid roll.

Chris: Sure. You get the Hello Fresh ad that plays at the beginning. Okay, cool. What are some other audio KPIs?

Anna: Yeah. So similar to OTT and CTV, frequency, that’ll be a main one. Also the reach of it, so you can have your impressions and then your unique reach. So those two and then just a couple other ones would be your device type. If you really care, you might only want to target people on smart speaker so you can have that higher audio completion rate, or you might want to only have mobile. And then the last one, just cost per listen. So the same as our cost per view, but just for a different metric name, basically.

Chris: All right, cool. So that really covers basically all of audio then. And then really the last main section that we would have varying KPIs for would be display.

Anna: Yeah, exactly. So the programmatic display, and we can also encompass native in here since it’s more of just like a dynamic ad format compared to a traditional static display ad, those are going to look more like our more standardized Google KPIs, or whatever channel you might be using. Microsoft audience network, for example. So depending on the purpose of your campaign, or goal, of course you might just have an awareness campaign where you’re really going to be focusing in on that reach and then potentially that click through rate. Whereas if you’re having more of a dynamic retargeting ad, or just retargeting ad in general, you might be more focused in on the actual conversion.

Chris: Okay. And then, just to quickly benchmark that, typically what we’ll normally see from a conversion standpoint is going to be much lower than a search campaign, but the impressions and the clicks and the engagement with that ad is going to be usually significantly higher.

Anna: Yeah, exactly. So across the board, a 0.1 is the standard benchmark for display. Obviously that’ll depend on industry and then your retargeting campaign might have more, but that’s where it lies in.

Chris: 0.1%?

Anna: Yeah.

Chris: Okay. Okay, so then to wrap things up, the main takeaway here is that when, when we are using programmatic advertising, thinking of the KPIs and the channel we’re on is going to be important to ensure the results are being reported on accurately and got to keep… Whoever the main client is in charge of managing these that they’re aware of what results we’re going to get and what they’re going to look like. Again, for example, we’re not going to get any clicks from an audio or a CTV campaign. And if we do, it’s going to be a very low number.

Anna: Exactly. So anytime we start to talk with a client about programmatic, being very upfront about what those benchmarks are is going to be your key thing to talk about for your key performance indicators.

Chris: Awesome. All right, Anna, as always, great to have you. Thanks for joining today.

Anna: It was lovely.

Chris: And, everyone else, please stay tuned for the next installment of our programmatic podcast series, programmatic out of home advertising.

Chris: Thanks for listening to the Getting Granular podcast. Be sure to subscribe so you don’t miss out on any PPC tips, tricks, or news in the digital marketing world. I’ve been your host, Chris Caesar. Thanks for getting granular with us today.