Offline Attribution with Krystina Rubino
Listen below, or grab it wherever you get your pods. Want to hear more? Each new episode brings in the best & brightest marketers to teach DTC brands about the art & science of questions. Subscribe for more, and let us know who you'd like to hear on the show.
Mitch (00:00): Welcome to Question Authority, where the best and brightest marketers teach brands about the art and science of questions. Today, we're asking about offline attribution with Krystina Rubino of Right Side Up. Krystina, thanks so much for coming on the show. Good to have you.
Krystina (00:33): I'm excited. Also, I'm liking that your audio sounds good because you have a balloon behind you. Respect.
Mitch (00:41): Yeah. I'm actually in a hot air balloon right now.
Krystina (00:43): Amazing.
Mitch (00:45): All right. So you are here today to talk about offline attribution and all that, that entails. So you've done that, I assume for several clients and in several iterations for all kinds of channels. So what's the TLDR of how you handle that.
Krystina (01:01): So I've actually been in digital and offline marketing for the last 15 years. I started out digital and then went offline, which is a different path for a lot of marketers who are like, "No, I want to go back to buying TV and radio because it's 1965." But what I realized was, I was getting really disillusioned with digital display and with a lot of the metrics surrounding it. I remember when viewability became a thing and I was just like, "Oh so no one's been seeing all of the things I've been buying. That's great. No problem." And I actually fell into podcasts about five years ago. I've spent time in-house. I've spent time working agency ends with Fortune 100, 500. A lot of what I do now is high growth tech, a lot of DTC some B2B and Right Side Up as a whole, as a growth marketing collective. I joined to start the offline practice because to your question, what I realized is that a lot of marketers think about offline as this big monolith.
Krystina (02:01): And this big tactic that they have to really just know what to do. And have a few hundred thousand dollars to spend at the outset at least. And have the right assets produced. And attribution's really scary because you don't have click-based and view through attribution. Last click is not an option for you, which frankly is one of the things that attracts me the most about offline marketing, because last click is a lie. And they didn't pay me to say that, just on the record. Yeah. TLDR is we use survey-based methodology. We work with everybody from Jordache, Calm, Stitch Fix, Brooklinen and a lot of the movers and shakers and podcast land and beyond. We help them with all sorts of channels. And we use survey based methodology across a vast majority of our advertisers. Just the best way to understand what consumers are really doing and what consumers we should be prioritizing, trying to get for customer acquisition efforts.
Mitch (02:58): Matt, you agree with that?
Matt (03:03): I think so. I think a question too, for you, I know we're going to dive deep into surveys, but for podcasts, what are the other common attribution methodologies that you see that, I would argue don't live up to the survey methodology. But utilizing discount codes, landing pages. I know there's definitely some black box attribution tech out there in the podcasting space that, I'm curious from your angle. What you think about the methodologies and how they are intertwined with each other. Just I guess, from your side of the table.
Krystina (03:37): And I should say, so our side of the table is we actually invest. All of our advertisers own their budgets in-house where a resourcing and a strategic solution for them. But we invested what, $50 million in podcast media in the last few years. So we've got a really good sense of what the channel can actually do. And to your point, the discount codes that you hear about, use code questions, get 20% off whatever the heck we're actually trying to hawk here. That is definitely a parody of podcasts. But we actually use that in combination with survey based methodology to get a picture of direct and indirect response. And that is the best and fullest utilization of the tech from a podcast attribution perspective. We like to say, I prefer vanity URLs just because I like to control the learning experience. I want to welcome people. Believe it or not when you welcome like, "Welcome, Joe Rogan experience listeners." There's actually a good bump in conversion from just that little, "Hey, I know you," people pick up on it and you're really trying to extend the relationship with a podcaster, that they have with their audience.
Krystina (04:51): There's a lot of trust built in there. So when they tell you to remember something, if it's a solid offer and somebody has a reason to remember it, you're going to remember it. And there's also a handshake thing, where people understand in that medium, that it is an ad supported medium. For better or worse. And the ads don't suck yet. I swear, I'm trying to do everything I can to make them not suck. Part of the challenge is that we are starting to see to your point, technology that's more from the digital end of the marketing spectrum, try to be applied here. And some of it's more successful than others. I'm excited for the advent of programmatic technologies applications to this medium. It's not happening right now. And in terms of measurement and attribution, we're seeing some of the same things. So pixel based methodology has started to become more prominent in the channel. The challenge here is, even if, "all of the networks accept it," many of the shows don't. When we go out to some of the attribution partners and say, "Hey, this is the media plan that we're running with.
Krystina (05:55): How many of these shows can you actually measure?" We only have anywhere between 20 to 40% of the shows accepting that methodology. And part of the challenge then becomes, okay, first of all, it's methodology. This is not one-to-one attribution. It's not possible. And even if it was, I don't know, I'm sure apple would have a couple of thoughts about that possibly. And that is also a great source of frustration for me. I'm like, "Oh, we're hurdling towards pixel based methodology as an industry. Anyone read the patch notes for iOS 14.5? Are we just going to pretend this is not happening?" This is the kind of stuff that frustrates me because I know we were chatting a little bit about Princess Bride before, and I'm sure people are sick of me using this analogy. But it's like that scene where he goes, "you keep using that word, and I do not think it means what you think it means."
Krystina (06:45): The methodology I laid out about extrapolation of survey-based methodology. That is the only tried and true way to scale this channel for performance marketers. Everything else is a newer thing that we're trying to use, to understand possibly more upper funnel metrics.
Mitch (07:03): Yeah. There is so much stuff that seems automated, that seems programmatic, and seems highly quantitative that actually has these little gnomes... usually know as junior underpaid marketing people or whoever... pushing buttons and moving things around to actually make the machine work. And I've seen that firsthand in places I've worked in.
Krystina (07:26): I was that 23 year who had way too much responsibility for budget. And you're like, "why did we do that to that poor girl?" And now I look back and I'm like, "My God." But it's a real thing. And I tell people all the time, I was at agencies for, I don't know, 11 or 12 years and I've seen it all. The stuff that, to your point, the stuff that you think is very regimented and data oriented. Data can be manipulated to tell a story. It is constantly.
Matt (07:56): Yeah. Mitch always likes to say, "The data doesn't lie until you start asking it questions."
Krystina (08:02): Dude, that is so good. And that's the manipulation that I see happen a lot. I'm a big advocate for clients and advertisers taking full control of their measurement methodology because I've never been a believer in having the company that's buying your media, control your attribution and measurement flow, end to end.
Mitch (08:23): Right.
Krystina (08:23): The big, bad wolf is not your best friend in that story. And it's not even intentional. Right? Tangent. I warned you guys that we were going to go off on tangents.
Mitch (08:33): Well, anytime there's a wild west, right? There's always going to be that like... even people who necessarily, I wouldn't say they mean well, but they mean innocently enough. Right? Where they're like, I discovered a thing and upon reflection, it probably wouldn't hold up. But I don't reflect as a general mantra. So I feel pretty good.
Krystina (08:53): I have no self-awareness. So no...
Mitch (08:58): Beautiful thing.
Krystina (08:59): Oh, man. Dude. Sometimes my husband and I are like, what? To go through life like that? Right?
Mitch (09:03): I wish we could still do lobotomies like they did in the '20s.
Krystina (09:07): Dude.
Matt (09:09): So our audience, the majority of brands that we work with today, don't advertise via podcast. You mentioned earlier in the conversation just around getting started and people are like, I can't touch this. It costs X to get started. What is that like obviously spending, it's not like Facebook or Google where you could spend 10 cents or a dollar or $2 and get a click. But what is that number and how would you go about introducing a brand that's, let's say doing seven, $8 million in top line gross revenue, trying to expand honestly, outside of Facebook, which is probably where 90% of their budget is allocated right now?
Krystina (09:48): Literally the most common thing we solve for. And also we happen to be a growth marketing collective that was started as a paid social and paid search collective because we all know the, "Traditional growth marketing tactics," that'll get you there. You can ask these channels to do what you have had Facebook and Ad words and all of these other tactics do. It's just that it's going to look different. You're not going to be able to use last click methodology.
Krystina (10:13): And Facebook makes it very easy for you to dump money into advertising and get an ROI or at least what you think is ROI... I was going to say, view through methodology, real generous one, you hit most of the known universe. We could get into that. But no, so it's a good, great questions. And we actually help a lot of the DTC brands we work with. I'd say like probably 75% of our current roster is DTC. And most of them are doing podcasts with us. At least on the offline side, I'm talking about particularly. We've had brands start with as little as $35,000. And as much as a quarter of a million to, we're about to launch a couple of campaigns that are spending four or 500 grand a month. So it really ranges.
Matt (10:58): Do you see podcasts... Is there typically an ROI right away or is it something that you invest in over a period of time? I know like Subway advertising, first couple of months you don't see anything. And then six months a year into it, it's like, oh, it's really starting to pick up. And we were going to turn it off and we didn't because we budgeted for it and now it's really working.
Krystina (11:17): Well. And that is, so iOS 14.5 has shoved everybody towards media diversification. Right? That is one of the trends that we're seeing this year. I've had product market fit for a long time with our practice. But what we've actually started seeing is just people are coming to us and already saying, "Hey, I'm hitting, diminishing returns in Facebook. And my next channel is going to be podcasts because it's a low production burden." In terms of being patient, there's actually a really good blog post that we wrote called Mind the Gap. There is literally an entire chunk of the campaign where you just have to hold hands and know that eventually it's going to come back around because let's think about it. So if you're listening to this podcast right now, you, human beings who are listening to this podcast right now. You probably are not listening to it on launch day. Maybe you downloaded it.
Krystina (12:07): Maybe you're streaming it. Maybe you've been saving it for, I don't know, your next run or something like that. Right? That could be two weeks from now. Because like, I don't know, I'm not exercising right now. But God speed if you are. Whenever that is, you're going to download and engage with that podcast. And that's the first point that you're going to experience this interview. And that's the first point that you would experience an Ad if we were embedding one in this episode. Right? So because of that delay consumption. And not only delayed consumption, but when are you downloading and streaming it? When are you consuming the actual episode? And then what if you're a brand that has a long purchase cycle? If you're a mattress company I'm not going to listen to the Ad and be like, "You know what? I'm going to buy a new mattress online tonight."
Mitch (12:51): I'm just going to light my current mattress on fire and go buy a Casper. That sounds great.
Krystina (12:56): That's I mean, I actually own two of them because of specifically, I'm not kidding, because of host endorsements from Howard Stern and Gary Dell'Abate. My husband heard them and was like, "You know, Gary loves his mattress." I'm like, yeah. I mean, it was a very funny thing. But now we love our Casper.
Mitch (13:14): Garry seems like the kind of person who has real trouble sleeping at night. Just more morally, physically, just all kinds of things.
Krystina (13:21): Possibly like stress-wise. Yeah. It's just like, and also that is actually just one endorsement where Howard proves the value of endorsement. He's been doing it forever. But that is the special thing that many pod-casters at all scale and sizes have with their audiences. So I recommend DTC brands start with podcasting. Also because it's an endorsement based medium. And when you're a brand, you're not Procter and Gamble. Right? Not everybody knows you the way they know Sharman. So how do you get over that trust and credibility hurdle with a consumer to say, "You know what, I do want to try a Quip toothbrush. I do want to order from Jordache tonight. I do want a Casper mattress." How do you get over that if somebody is just exposed to your brand for the first time? The answer is, you're borrowing the credibility from the host.
Mitch (14:10): Right.
Krystina (14:10): And in terms of how to set up podcasts so that it is successful. I like to tell people, you should always, any offline channel period. Whether it's anything from television, to SiriusXM, to terrestrial local radio, to podcast. You have to contextualize that test spend against the rest of your channels. So you're going to see down funnel activity from doing any broader prospecting advertising. Right? That's just a fact. But you're not going to feel it if you don't spend enough to have people searching for your brand on Google. So then the question is, "Okay, so you're spending 600 grand a month in Facebook. Why did you think you could spend $2,000 on a podcast? Why did you think this was going to be a significant?" Even tell you then to say, "Oh, I can scale this channel to a meaningful part of my marketing mix." I'm all about contextualizing the size of the test spend.
Krystina (15:05): And because we don't work on media commission, we're a retained practice. We can literally just set the right spend. And so to your point about some of the best practices. Some of it's basic and some of it's just good market research design. But the challenge is when we work DTC. A lot of the times, you're the first marketer that's hired. So like you don't have the luxury of a market research person. I used to work with brands like Advil and Robitussin who had the luxury of having all the market research people that we could ever want. I never had to design a question back then, what are you kidding me? I didn't even realize that there was an actual art and a science to it.
Krystina (15:40): But when you don't have those people around, what are the basics? It's like, okay, well, we should be anchoring our choices for our survey in the channels that we are using now and are intending to use. Because we want to make sure that we get a solid baseline so we can do baseline removal. If can't do baseline removal then you're going to be so surprised, that when you have the survey live for a month, and then you start to see when you turn your podcast spend on, that 1% of your conversions were already coming from podcasts. People might've heard your founder doing an interview, or may have confused you with one of your competitors or something like that. There's a million reasons that people get confused when selecting choices. And then also just like common sense stuff. Like word of mouth.
Mitch (16:28): You mentioned word of mouth. It's hilariously forgotten by so many DTC brands and marketers who just think that you push Facebook buttons to do marketing. And it's like, any decent marketer knows this is arguably, the most important channel. And granted the most challenging. But to your point about setting that baseline pre podcast advertising or whatever channel you're diving into. That's a great example where I've seen brands where this has happened. Where if you set that baseline for word of mouth and you start running podcast as.
Mitch (17:01): You see word of mouth jump up, because it's like, well, word of mouth. Its like, to your point about the testimonial. Right? It's like, well, word of mouth in the front of my mind. And I guess in the back of my mind, if you ever really drilled in the word of mouth is Gary Dell'Abate. So it's not my friend, Gary. It's the podcast I listened to that has Gary on it. But I feel like it's word of mouth. So understanding that Delta is crucial. Right? If you're going to spend 20, 30 or 300 grand on podcasts and see like, oh, there was like a 20% jump and word of mouth post podcast spending. That makes sense.
Krystina (17:36): I also try to ask advertisers. I'm like, why? This is how most of the brands and specifically podcast advertising, we're getting meta and talking about podcast advertising and a podcast today. But specifically when it comes to that, it's like, this is how everybody from ZipRecruiter, to Mandy's, to Door Dash, to Calm had scaled this channel in the past. So why are you trying to get fancy? And why are you trying to do it differently? Yes, when you... The 201-301 version, when you're spending five, six, $700,000 in the channel every month, and you can take those big bets and make those big swings. And overall the channels profitable for you. And you know that to your bones. Then you can get fancy. Then we can look it up or funnel metrics. Then you can layer on pixel based attribution and see if it actually is worth the pixel implementation itself. But for the rest of it, go back to basics because if you've been able to scale paid social, you should be able to scale at least a couple of offline channels.
Krystina (18:48): If you know you have product market fit and you've been able to establish your basic growth metrics and you know, your LTV and AOV and all of the other acronyms that we love to use. If you know all of that stuff, you'll be fine. You just need to work with people. Whether it's hiring someone. Whether it's a consultant. Whether it is engaging an agency. Whatever it is, you just need to work with people who've done it before. And then work with them to adapt how they do it to your business. Don't try to go it alone. And also please, if you take nothing else away from this, please don't buy one episode of a podcast and thinks that that is going to be indicative of how the channels performs. Because I like to say, so quick and dirty. Somewhere between $50k to $150k for a seven to 10 week period, is what I usually recommend for 85, 90% of the brands we work with, to test the channel. There's outliers in either direction.
Krystina (19:44): We've started some brands as low as $30K because they had baby marketing budgets and it was an appropriate spend. And you can get a really good podcast task going for that. But make sure you're testing multiple episodes per show. No fewer than two. Ideally three and sometimes four. Just depending on the complexity of your message on the show itself. There's so many different factors to your question earlier, Matt, about selection criteria. So many different things go into how we flight media. And the last part is, make sure you give yourself a few weeks in between integrations because of that delayed consumption that we talked about earlier, you tend to see that media reach and frequency in this channel accumulate over time. And that's the last part of it. Reach and frequency cume, is not a metric I thought I'd ever talked to growth marketers about. But guess what? That's what it is. It's just the reach and frequency of media exposure. So we can get fancy and call it whatever we want as growth marketers. But at the end of the day, it's really just about measuring consumer response to media exposure.
Mitch (20:50): Yeah. And to conclude, I think you made that great point just a couple of minutes ago about, take the wind, girl. I'll take the wind from all the Facebook ads that you did and everything. It's like, oh, we found product market fit. We have our numbers on AOV and LTV and all this other good stuff. Great. That's the kind of foundation that will make you super successful in offline media or anything of that nature. Right? Is like the stuff that is maybe harder to ramp up from zero when you don't know those things. You got the benefit of using something like Facebook.
Krystina (21:20): There's a bunch of ways to get started in these channels. I like to say, if people are interested in reaching out to me and talking to me about working together, I'm just always down to have that conversation. We've also published a lot of growth marketing guides and content on how to do these channels. So if you head to RightSideUp.com and you go to our blog, you'll see a bunch of stuff that I've done in the past. Like podcasts, webinars, both that my and my team have done. And otherwise, I love to be stumped. If you're navigating these channels and you're like, "Damn. I don't even know how to make heads or tails of this data." That's what we do for a living.
Mitch (21:53): Well, thanks so much. This is very insightful. As we all knew it would be. And very entertaining. Some of which is not going to make the cut for the actual episode but-
Krystina (22:01): I prefer that. That's perfect.
Mitch (22:16): That's going to do it for us today. Thanks for listening, subscribing and rating the show. If you're digging what Krystina was laying down, check the whole team out at RightSideUp.com. Do you want to chat with Fairing Labs? Head on over to Fairing.com. See you next time.