Today in Digital Marketing - Deep Dive: Finding and Ranking Potential Brand Alliances

Episode Date: February 27, 2024

Sometimes, one of the tasks of a marketing person at a company, is to propose partnerships between their brand and another. This offers the opportunity of joint advertising campaigns (like Coca-Cola and McDonald’s), cause-based alliances (like Target and UNICEF), or even bundling products together — like the streaming deals that include joint Hulu and Spotify subscriptions.But finding those partners isn't always easy. And determining whether your customers would be thrilled with an alliance or offended by it should be top of mind.What if there were a measure of brand partnership potential? A score that would tell you which other organizations would make for great alliances.That's what Pankhuri Malhotra set out to invent. She and her coauthor have published a paper in The Journal of Marketing called Leveraging Cofollowership Patterns on Social Media to Identify Brand Alliance Opportunities..GO PREMIUM!Get these exclusive benefits when you upgrade:✅ Listen ad-free✅ Back catalog of 20+ marketing science interviews✅ Get the show earlier than the free version✅ “Skip to story” audio chapters✅ Member-only monthly livestreams with TodAnd a lot more! Check it out: todayindigital.com/premium✨ Already Premium? Update Credit Card • CancelMORE🆘 Need help with your social media? Check us out: engageQ digital📞 Need marketing advice? Leave us a voicemail and we’ll get an expert to help you free!🤝 Our Slack⭐ Review usUPGRADE YOUR SKILLS• Inside Google Ads with Jyll Saskin Gales• Google Ads for Beginners with Jyll Saskin Gales• Foxwell Slack Group and CoursesSome links in these show notes may provide affiliate revenue to us.Today in Digital Marketing is hosted by Tod Maffin and produced by engageQ digital on the traditional territories of the Snuneymuxw First Nation on Vancouver Island, Canada.Our Sponsors:* Check out Kinsta: https://kinsta.comPrivacy & Opt-Out: https://redcircle.com/privacy

Do you have business insurance? If not, how would you pay to recover from a cyber attack, fire damage, theft, or a lawsuit? No business or profession is risk-free. Without insurance, your assets are at risk from major financial losses, data breaches, and natural disasters. Get customized coverage today,

starting at $19 per month at zensurance.com. Be protected, be Zen. Hello there, it is Todd. I am still on vacation, Be protected. Be Zen. at a brand is to propose partnerships between their brand and another brand. This offers the opportunity of joint advertising campaigns like Coca-Cola and McDonald's, or cause-based alliances like Target and UNICEF, or even bundling products together, like the streaming deals that include joint Hulu and Spotify subscriptions. But finding those partners isn't always easy. And determining whether your customers would be thrilled with an alliance or offended by it should be top of mind. What if there were a measure of brand partnership potential? A score that would tell you which other organizations

would make for great alliances? That's what Pankhuri Malhotra set out to invent. She and her co-author have published a paper in the Journal of Marketing called Leveraging Co-Followship Patterns on Social Media to Identify Brand Alliance Opportunities. I spoke with her earlier. So we were specifically looking into Twitter and we were trying to like look into the engagement patterns

of the brand audience of a certain set of brands. And then we realized that, hey, there's a pattern here. So if we look into the co-followership patterns between certain groups of brand, some groups of brand have very strong co-interest patterns than the others. So we saw that and we were like, okay, there's definitely an opportunity here. And then we started collecting data. And what we found was we found these like clusters of brands in these networks. So we are finding clusters of brands like Tesla connected to technology brand.

We were finding brands like Nike connected to NFL and NBA. I mean, based on what people were like, based on how people were engaging with them. Yeah. Connected how? What do you mean? Like this was grouping hashtags together? How did you link these companies with topics? Followership activity.

So if I am following Tesla and Budweiser together, there will be a link between the two. I see. Yeah, and what we did was we looked into like the entire brand audience of all these brands. So like Google, they have like 20 million followers. NFL, they have like 25 million followers. So what we did was we looked into these common followership pattern. So it's not just one person who's co-following these two brands together. These are like millions of folks. So there was this range in the co-interest

patterns between certain groups of brands. And I think that range was very, very interesting to us because in some way it was telling us that, hey, there's a potential brand alliance opportunity that the managers may not be aware of. I see. So the practical application of this is a way to identify potential partnerships between brands based on them having similar Twitter followings? Do I have that right? Yes, yes, exactly. Exactly. So that's the objective. And I think one of the advantages of our approach is like, we look into brands from different categories. That's the objective. So previously in marketing research, there have been studies that have tried to identify co-branding alliances, usually within a single category.

So there's been a scalability issue. So our focus was slightly different. We were like, okay, you already know that two brands in the same category, they're competitors. People might use them at the same time. But what about brands in two distant categories? Nobody talks about that. Like Starbucks and Spotify, they're like two seemingly unrelated brands, but they recently had this whole co-branding alliance. And the fact that they did have this alliance helps them to like cross promote and grow their consumer base. So the focus was mostly on brands belonging to different categories. So if I'm a technology brand, can I partner with someone in the sports category?

If I'm a beer brand, can I partner with someone in the airline category? So that's the objective. Right. And your research, of course, you had access to big piles of data from Twitter that you put into, I presume, a large database, and it spat out these alliances or potential alliances. How does though an average digital marketer who is managing a small brand, for instance, try to identify a potential partnership? How do they even like download the followings? So it's publicly available

data. So that's the good thing. And through this paper, so like we're providing them with this automated tool. So we've given the entire methodology. We've written the code for them. So this is like a tool that they can like directly replicate and use. Oh, I see. Okay. Yeah. So like we're providing them with this new methodological tool that, hey, this is a brand

network. This is how it's created. So we have the entire data. We have the entire code and they can use that to like create their own brand networks and see, okay, how is my brand audience interacting with other brands in the sports category or like any other category? Is that online?

That's a tool online? It's not really a tool as of now. It's in the paper. Like we've like, I mean, supplied all the material and everything, but at some point we do want to come up with like dashboard that we can supply to managers. But right now, it's just code and the entire methodology. So it's sort of like if you've got a developer, these are all of the steps you would need to take to produce your own version of this. Yes. We're not there yet. We're not there yet.

What is brand transcendence? I saw that mentioned in your paper? Oh, yeah. So like in this article, we also come up with this new construct called brand transcendence, which is essentially defined in the context of brands belonging to different categories. So brand transcendence tells you how it tells you the extent to which your followers overlap with those of other brands in a new category. So for example, if I'm Heineken, a beer brand, how do I transcend into the sports category? And we come up with this construct that's based on the extent to which my followers, for example, Heineken, they overlap with all the other brands in the sports category. So we come up with this construct. And the good thing about this construct is, again, it's an automated, it's an automated construct that tells you not only about

your cross-category connections in the sports category, it also tells you about your cross-category connections in the airline category or in the dining category. So you can compare your transcendence and see which categories are most suitable for me to extend in the future. And is it a measure? Is it like a score system of some kind? It's a score. So it's like from negative one to plus one. So if your values are closer to plus one, that means that you're highly transcending into a new category. So, for example, like in our paper, I think we talk about the transcendence for car brands. And, of course, we see brands like Mercedes, which have very strong transcendence into categories like technology, luxury, retail, suggesting that extensions might be possible for Mercedes in these different categories. Then we also have brands like Toyota and Ford, whose transcendence to different categories is not very high, but their own centrality in the

automotive group is very, very strong. So most of their followers, they're interacting with other brands in the automotive category. So they're not transcending as much, but they're very, very mainstream within their own category. And same for brands like Tesla, like they're not so central within the automotive group, but their transcendence to categories like technology is super, super high. So there's a score that we're giving to each of these brands, depending on how their followers are overlapping with the different categories. Did you discover any sort of commonality between brands that were more transcendent than others? Like is there, if someone is a brand manager or a marketer inside of an agency, and they're thinking about their clients, they're thinking

about your work, is there a way for them to determine whether or not they or their brand is more transcendent than others, or is more likely to be more transcendent? In other words, to have more of these potential opportunities? Are there things in the brand itself that make them more transcendent? Oh, well, that's a very, very good question. So like we did not look into the causality part of it as of now. So as of now, like it's like a BI tool. It's a business intelligence tool that, hey, you have this network and using this network, you can get a score. And I think with transcendence, there are two things. So you can have high transcendence

because of the fact that you might have done some form of like marketing campaigns in that particular category, for example, Red Bull and FIFA. So like Red Bull is like super connected with FIFA. Why? So they've had like marketing campaigns before together. So the transcendence in this new category is very, very high for Red Bull. So it could be because of two things. First, of course, because of the previous marketing actions of the company that can eventually impact its transcendence in this given category. Second, it could be entirely something new that managers are not even aware of. And we saw some like, so we tried to like manually see why this could happen. So we had like instances like, I think Pepsi and Budweiser

where we're like, okay, there's a very strong co-interest pattern between the two. And what we realized was, okay, there has been a campaign between the two before. And that's why we're seeing this pattern. But then we also had brands like beer brands, like Sierra Nevada, that were transcending a lot into the sports category Nevada that were transcending a lot into the sports category. They were transcending a lot into the technology category. And we were like, okay, why is this happening? There's not been any marketing campaign between Sierra Nevada and Microsoft, but still their followers are like super engaged with each of the brand. So it could be either of the two, either your marketing actions, or it could be something that the

managers are not even aware of. Do you have business insurance? If not, how would you pay to recover from a cyber attack, fire damage, theft or a lawsuit? No business or profession is risk free. Without insurance, your assets are at risk from major financial losses, data breaches and natural disasters. Get customized coverage today starting at $19 per month at zensurance.com. Be protected.

Be Zen. I'm trying to figure out how a marketer could do this in sort of a real-world situation. You're right. The people who are following each account is public data. It's just not easy to get out of it.

You've got one of two approaches. You either hire a developer or have a developer on staff, get access to Twitter's API, and then pull the data that way. Not many brands have that. The big ones do, but do they want to spend their developer time on that? Or you hire a team of people off of a freelance site

to manually enter everyone's handle on the following of both accounts. Like, I'm trying to figure out how does someone determine this if they don't have access to, you know, the API or developers that can program it? I think it's like, the thing is, the data that we're using, it's not private. It's like publicly available. So like, you can be from the academic group. You can be an industry practitioner. It's available to everyone.

But I think the skill set part is something that is something to consider. The good thing about our approach is that it's automated. It's scalable. We're providing them with the method. But I think, yeah, I mean, that's a good question. I feel like most of the companies, I think even smaller companies, they tend to have like marketing analysts or like data analysts that can like replicate this approach for the focal brand if they want to. But as of now, we don't have like a dashboard that brands can go and get a transcendence score. We don't have that right now.

Yeah. So, yeah, but that right now. Yeah. So yeah, but that's something to consider. But if someone did have a developer, they could use that tool to get that. Yes. So like we outlined the methodology, like all the steps on how to calculate your transcendence. We also actually validated it with a study on Amazon Mechanical Turk. So like we did a survey. So I mean, one of the overarching questions for us was, okay, the social media patterns are telling that there's a link between these two brands,

but does that link really exist? So like we did do a survey and we gave these people like sets of brands and we asked them that, oh, will these brand pairs go well together or not? And we did find some very interesting pairs that we were like, okay, I mean, like Audi Intel, like, Twitter was telling that, hey, there's a strong connection between the two. And somehow, even the survey people did rate these two brands very high. So like, so that was like a good, good, good correlation with, I mean, between the two constructs. You mentioned it to score negative one to positive one. I could I ask you to reveal sort of how that formula works? Like, is it not, does it need to be that complicated?

Could it not just be, you know, if there are 5% of the followers between both, or, you know, that are common between two accounts, that's good. If it's 1%, it's bad. If it's 20%, that's amazing. Can it be that simple a formula or a calculation? It could be.

And actually, that's the first step that we do. Like, the first step in our analysis is to find the percentage of your followers that are interested in the new brand. That's the first step. But then I think what we have to eventually do is not just do it for like one brand. We want to do it for the entire category. So we extend that formula. So like for Heineken,

like the transcendence is not just to NFL. We also want to find its transcendence to Olympics. We want to find its transcendence to NBA. So like in the entire sports category. So we start with the percentage and then we extend it to the entire category. And the reason why we have like negative and positive value is so like certain categories, they have like some form of like intrinsic connections with other categories. For example, sports and beer.

So like on an average, you will find that sports audience, they're also interested in beer. So what we do is for each of these brands, we compare their transcendence with the category average. Is your transcendence above what your category has in general or not, or is it below? So that's why we have this negative and positive. So are you doing better than what the rest of the category is doing, or are you doing lower than that? What surprised you the most about this research? I mean, some of the connections that we saw, we were like, okay, this, I mean, we never even thought like, okay, that there would be a connection between like these, these, these two types of brands.

Can you give me an example of the, of two that surprised you? Sierra Nevada, like I'm going through some of the findings. So like Sierra Nevada with like Netflix and Microsoft, that was interesting to me. So like Sierra Nevada with like Netflix and Microsoft, that was interesting to me. So like Budweiser was more with beer and food fans. Like we can see that, okay, most of the people are interested in Budweiser. They're interested in Taco Bell.

They're interested in Pizza Hut. But with Sierra Nevada, it was not like that. Their audiences are there more into like Discovery, Microsoft. So the complete contrast that you see in transcendence for competitors within the same category, it was very, very interesting to us. Same for Tesla, actually. So like Tesla had a very, very high transcendence

to the technology category as one would expect, but its transcendence to luxury was not really high compared to like Mercedes or BMW. So people like think of Tesla as more of a technology brand than a luxury brand. I know you didn't study this, but do you have a hunch as to why? I do think it's like the brand concept that plays a part here. Like Tesla has always been a tech brand. They've promoted themselves like that compared to BMW, which is more of been a luxury brand. I think it has something to do with the way these brands have promoted themselves or positioned themselves in their audience's mind.

So I think it definitely has to do with their perceptions in consumers' mind. It sounds like it might be a really good tool to gut check your own brand against what you believe your positioning in the market is. So if you believe, you know, like if you're Tesla, obviously the people who are behind Tesla believe that they are positioning their messaging as the luxury brand, but maybe if they had access to your tool and they ran it through,

they would discover actually the market isn't responding to us in the same way. I wonder if this might be a tool for brands to kind of gut check their own position in the market. Yeah, I mean, that's one of the applications that we talk about. So like, since fluctuations between a brand and a category, they can change over time, we do a dynamic analysis. So we compare our findings, like for 2017 and 2020. So there is like a huge change that happens. Like for some brands, the transcendence goes up. For the other brands, the transcendence goes down. So for managers to know that, hey, connections are being formed

between a new category, it can help them to identify potential co-branding opportunities. At the same time, if they get to know that critical associations to some categories, they are waning or they're going down, it's like a check for them that, hey, we need to do something about it. So yeah, it can work both ways. Yeah. What made you want to study this? Overall, I'm like a computational social scientist. I'm very, very much interested in like learning and exploring on how users interact with brands on social media. So, I mean, that was one of the reasons to see if I can leverage these digital footprints of users on social media platforms to come up with a tool that marketers

can use easily. And co-branding for me, I think it was the most straightforward application that we could see right there. I think there could be more applications of it. We're still in the process of identifying those applications, but co-branding or brand alliance was the most straightforward that we could just simply see based on the co-followership patterns. Do you have plans to make that tool available publicly? I mean, like in, in, like in working fashion, like where you could go to brand transcendence.com and, you know, it would be there for people to use. I would love to, I mean, I would love to do that. I mean, that's the ultimate goal that, you know, we can come up with like this automated dashboard,

which is dynamic, right? So like marketers can see, okay, this happened in January. My transcendence was so high with this brand or with this new category. Then in February, it absolutely went down and it can happen because of multiple things, right? The brand's own promotion, what your competitor is doing can also affect that. So for them to know, like for them to have this automated dashboard, I think that would be very, very interesting. And maybe that's the next step for us. I hope it is because so much work gets put into science and marketing research, and then it often just sort of stays at the theoretical level.

And I know many marketers, I think, would be very interested in using this. Thank you for sharing this with us. It was very interesting. Thank you for having me. Pankri Malhotra's paper is called Leveraging Co-Followship Patterns on Social Media to Identify Brand Alliance Opportunities. It's in the Journal of Marketing. It's the season for new styles and you love to shop for jackets and boots. So when you do, always make sure you get cash back from Rakuten. And it's not just clothing and shoes.

You can get cash back from over 750 stores on electronics, holiday travel, home decor, and more. It's super easy. And before you buy anything, always go to Rakuten first. Join free at rakuten.ca. Start shopping and get your cash back sent to you by check or PayPal. Get the Rakuten app or join at rakuten.ca. R-A-K-U-T-E-N.ca.