How marketing spreads / Data for Bluffers #11

11 April 2022

Why do some campaigns work and others fail? How does your product type influence how you should target your marketing activity?

In this episode, Tom and Ed discuss how opinions and ideas move between people and how you can harness this to have more impact from your marketing.

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

Tom

Welcome to another episode of the data for Bluffer’s podcast. This week, we are looking into the data and science point, how opinion spread, which is critical. If you’re trying to get people to listen to your marking, you’re trying to get your marking to travel and, and reach further example. We talk about a lot is that our brains filter out the unfamiliar. So if somebody doesn’t know about your brand, they’ll subconsciously filter it, AKA ignore it. So we’ve always got a slight chicken and egg scenario in that you have to be in someone’s head before they’ll acknowledge you. And as the vast majority of people get their awareness from their social groups, it makes sense to target these groups. However, how opinion spread two. And within these social groups, isn’t a one size fits all approach. So today I really wanted to dig into this and understand how marketing and growth teams can use this insight to better reach and kind of convert new customers. So ed, how do opinions spread?

Ed

There’s no real simple answer to that. Yeah.

Tom

I’ve come to expect that now.

Ed

Well, it’d be a very quick podcast.

Tom

Exactly. Yes.

Ed

When mathematicians and scientists are studying this, they, they build models that sort of replicate the processes that go on in society and in friendship networks, to understand how those, how opinion spreads and those models, we can sort of categorize into three broad groups. There are debates amongst scientists about which group is, is most realistic, different people have different preferences for that, but also it’s very much dependent on the situation. What is the marketing message or the adoption behavior? So, so taking up a, you know, the use of a product or an activity like smoking or joining a fitness club for those different scenarios, we can talk about different processes that are most appropriately model what’s going on. So at one end of the scale we have, what is really kind of simple copying. You can imagine this, that you sort of look at one of your friends, you like know the para trainers they have, and then you copy that behavior. And this is very similar in some cases, identical to the spread of diseases

Tom

Makes

Ed

Sense. So Quinn, we can sort of think about this in the, in the disease space. So the, the idea of, of almost contagious behaviors spreading through the pop. Okay. Slightly more complex than that. Okay. We can talk about complex copying mechanisms. So here it’s where there’s a more complicated process that leads to you copying the behavior around

Tom

You. Okay. Give, give me an example there.

Ed

So the most common example of this would be like a, what is called a threshold copying. So that’s where it’s not enough that one of your friends is doing something, but if two or three of them are,

Tom

Yeah, yeah. Okay.

Ed

Then you might adopt that behavior. So maybe something like a, you know, like a new bank account, for example. Yeah. You’ve never heard of Monzo. You are not gonna be the person that signs up first or second, but once a few of you, once two or three of your friends have had it and their money’s not gone missing. Yeah. You, you, you start to think like, oh actually this is something I might do. Yeah.

Tom

Yeah. And I, that was a, that, that’s a great, great use case. That’s exactly what, you know, when I, I used Monzo for a while and that exactly it, I, a few people started using it and then actually I think three or four friends, um, in the tech space were using it and I went, oh, well, okay. It’s good enough now. And, and, and then I dived in.

Ed

I exactly. And, and I think that’s, that’s very much, uh, yeah, that’s sort of, you know, you get that authority almost about the product from how mean multiple use cases around you.

Tom

Yeah. That already, that’s really, that’s really interesting to think about actually that, you know, we, we talk a lot or I talk a lot certainly about how, you know, we will take, most people take a steer from their, their social groups and people around them, but, but already actually yeah. Splitting it into kind of simple and complex is, is quite a good way to break it on. Cause it really, and I think as you said at the beginning, what, what type of product or whatever, what your product or your service is, is gonna fit into different camps potentially. Um, is, is that it is, is, are they kind of the core two or do we, can we break it down further?

Ed

There there’s one more, which is, uh, or there’s sort of the other end of the scale, are it say from simple copying sort of majority decision making where you’re trying to decide on a, on a purchase and you kind of take a survey almost of your friends and take the majority decision. Correct. Okay. Now this is often, uh, I’d say there’s two, two sorts of products where this is probably the best way of thinking about things. One is, is in sort of connected products. So social networks and things like that. The other example, and this is really where these models sort of developed is in the case where you are making a choice out of a discreet small number of different options.

Tom

Okay. You got a couple, yeah. A couple of things on the table and you work out who’s and what, and go, I’ll go with it, go with whoever’s doing the most.

Ed

Exactly. So, in, in, in terms of how the model was developed, you would, the, a lot of the opinion, formation models are developed to model voting behavior where you have two parties and you’ve gotta choose between them. And if you don’t really have a decision, you might survey your friends and say, what are most of my friends doing? Okay, I’ll vote Democrat or I’ll vote Republican

Tom

Broadly, I guess, you know, we’ve got these, these three areas we’ve got, you know, simple copying behavior, complex copying behavior, and then surveying, we’ve kind of got these three types you’ve you’ve described, I guess I’m now mentally trying to build out a matrix that says, well, what, what characteristics of my, my product or service, what sort of attributes I guess, um, will we look for, to put into the, to each of those three buckets

Ed

Kind of there there’s a few key characteristics of a product that, um, really determine what is the most appropriate way to think about its, its sort of behavior. And I think in terms of like for a simple copying, we are often talking about lower cost products or products that are kind of naturally low across, but also potentially have a very high sort of refresh time. So disposable products, because you’re only G you’re literally just copying off one friend, you are really not putting a lot of effort into that decision.

Tom

So a new, a new coffee supplier, I buy my coffee from, you know, roasters, ABC, but you, them from someone else I might think, well actually, you know, ed says he enjoys it, so I might as well try

Ed

It. Yeah, exactly. So EAs E easily sort of reversible decisions as well is one way of thinking about them.

Tom

Yeah. Okay.

Ed

Yeah. Often lower price or with the ability to, to go back to some extent also an product. So it doesn’t necessarily have to be replacing something that you are already doing is, is the equivalent of like stopping somewhere for coffee when you normally wouldn’t on the way to work. Right,

Tom

Right. Yeah.

Ed

Okay. By low cost, I mean low cost to the, the consumer in terms of effort and investment. And it’s not a decision they’ll be stuck with for a long time.

Tom

Okay. So it probably for I’d say not, not none but less B2B products would fit in this category. You’ve probably got as proportion, more B2C products.

Ed

Yeah. I think, I think B2C products in the B2B space, I think here, we really probably are talking about consumables is, is the most, okay. You know, obvious example of this, you know, your printer, paper, you a printer rink potentially also in terms of low cost, like companies that have free trials. So short term subscription services, you can turn on and off, or actually that is a good transition into what sort of products fit into the sort of complex copying space. So in complex copying a lot of the time these, so the, to recap like the example we used as the threshold model, so where two, two or three of your friends, um, or if you’re a business, your peers or your, you know, not work colleagues, but people, you know, who do similar roles at other businesses were kind of adopt a product. And then you adopt that, uh, adopt that product too. And this, this is where I think a lot of lower cost B2B product set, particularly say, I’m thinking about like SaaS products here.

Tom

Yeah. You, you, you know, a couple of people that use it, you know, you you’ve, you’ve waited for a few people and then once you’ve heard a few, then you, you kind of jump on

Ed

Board. Exactly. And they, they may be. And I think the reason they fit really into this space is that there is a, there’s a lower, they might not be, you know, particularly high cost for the business, but there’s a time cost in adapting to them using them, uh, you know, in the SaaS space and it, and that’s why you want to know that effort

Tom

Is worth. Yeah. And I guess another one that fits in here is probably around trust and security because, you know, I think for, for some of these products and services, you might be putting sensitive data in, or they might be performing security services or, or whatever it may be. But I think, you know, as you move further up that trust scale or more trust is needed, um, they fit more snuggly into the complex copying space. Is that fair?

Ed

Yeah, definitely. I think, you know, as I say like this, this is really about building authority around the product, for example, for startups or startup products, right. It, this is always kind of a big fear when buying them that they’re not gonna here next year, right? Yeah. Yeah. And the more you have to integrate into them, the bigger that fear is. And so knowing that other people are using them is just some sort of benefit. Yeah. We were talking about this, uh, I think it was on the last episode about people gaining authority by putting, you know, logos on their website. People use that as almost a shortcut to this sort of called Plex copying. Right.

Tom

Yeah. Okay. Yeah.

Ed

Yeah. So instead of copying your friends, you are kind of copying the companies on the website. Yeah. And I mean, one thing we do know is that’s nowhere near as effective when it comes to social behaviors.

Tom

Yeah. We talked about it last time. Right. The, I, the ideal would be that when you land on a website, I, I go on there. It says, your friend ed uses this like, oh great. You know, it’s

Ed

That, the other thing that fits into this space is very high value products and, and potentially like long term investments

Tom

Yes.

Ed

Where you need the authority. So, um, a good example in the B2C space would be say, like getting solar panels. Yes. Something we’ve spoken about recently. Right. So yeah. You know, I, now I’m in the position where I, now I know three people have had solar panels put in recently and that’s very much gone to the point of gone from the point of like, oh, they’re probably very expensive. It’s probably a lot of hassle to, oh, actually it’s something I could consider.

Tom

Well soon, no, if any, uh, solar panel installing businesses, listen to this, whether you get some, uh, barrage with phone calls knowing you are in the market. Exactly. And for me, like how does that differ between? Cause for me, complex and survey seem quite similar. Like I, I get the fact that survey is, is more a majority piece, whereas complex is not, you know, if two or three of my friends have, have something I might buy it, but for a survey it might be nine, 10. Do these kind of attributes, are they similar or are there kind of other ones we’d look at more if we were looking at the, the survey end of it.

Ed

So I think, I think for survey the most important feature that categorizes a product as being in the sort of survey mode is when it is directly replacing another product. Yeah.

Tom

Okay.

Ed

No. Now we talk about Android versus iPhone, but historically we could also say that when smartphone first came out, that people were kind of waiting for a majority of their friends to have smartphones before making that switch over to smartphone themselves. So when we talk about replacement, we, we often can focus on like highly interchangeable items. You’re very unlike, well, you’re not gonna buy both of them. Right. And that that’s the important factor here is, is that is having both is not an option.

Tom

That’s a good way to frame it. Yeah. Like that, like that

Ed

Now, obviously, and this is sort of, so that we, we haven’t bought up yet, but is really important. Is that the way that people interact with products is not the same for every person

Tom

And probably not the same for every product as well. They might be the early adopters for some products, but actually for other products, they’re gonna wait and, and, and see, see what happens.

Ed

Exactly. So like the smartphones, very good example where, you know, if everyone’s away for a majority of their friends to have one, no, no one would have it. Yeah. But there were some people who were early adopters in that who would like want to be the first person to have one. Yeah. Then there’s some people who are sort of complex copying. So they would, you know, once a couple of their friends would have one, they don’t wanna be left behind and then finally get the people who’s like, well, I one now, because I wanna be able to talk to my friends on WhatsApp and things like that.

Tom

But if we, I guess if we summarize a bit there, you know, we’ve the core things we were really looking around. There were price, refresh time complexity, or, or security or trust, that type of thing. Looking at those things helps you put yourself into those three buckets of whether it’s a simple copying thing, you know, trying a cup of coffee or maybe something a bit more complex. How do we like, do we have to think differently then about how we use them? So, you know, if I’m, if I’m, let’s say I’m the coffee supplier and I, you know, I think my product falls into a more simple copying mode, whereas, you know, maybe I’ve got some security software. So I think I’m pro probably more complex copying. How do I have to think differently about that? If I’m, if I’m trying to, trying to get my message to, to spread and, you know, reach, reach more people and, and ultimately yield more customers,

Ed

If you’re in purely a simple copying world, then the key factors you, you, you want your, you want your product to spread as far and wide as possible because it, it doesn’t matter who, who is, you know, one of your users, they can pass it onto their friend. The truth is that very few products actually fit exactly into that model. There tends to be, you know, there’s brand building and, and all the word of mouth and, and things like that. And, you know, like soft recommendation that we’ve spoken about, the more complex, or like towards the survey end of the spectrum, they’re, they’re sort of harder, harder products in a way to, to spread. So the, the simple example is in a threshold model of two is no good spreading your product as far or far and broad as possible because no person then has two friends that are using your product. And therefore you’ve got no one to be the next Okta.

Tom

Yeah. Okay. Yeah. Makes perfect sense. And

Ed

That’s, that’s what you wanna try and avoid in that situation. Yeah. Now luckily, the way that we connect up in our friendship groups has done a lot of the work for you as a marketer, friendship groups tend to have what in the network world is called clustering. If two people are your friends, they are more likely to be friends with each other than two random people in the world. So if you think about it in terms of connections, it’s about forming these triangles of connection. If you can target those groups, it makes it easier for the complex behaviors to happen,

Tom

Cross-pollination going on between them.

Ed

Exactly. And at the, and at the extreme end of the scale, we can talk about cliques, which are sort of fully connected groups of people where everyone knows everyone. And in those situations you get very like for example, uh, with, uh, survey behavior, you get very, very strong set behaviors to the extent that if you are in a clique, it’s very hard to change your behavior in a survey model, because you’ve got 10 friends in your clique, you all think the same thing and every survey you do of your friends, they’re gonna dominate that really what it comes down to is depending on yeah. Where your product is sort of either helped or hindered to a certain extent by this clustering that naturally exists. Ah,

Tom

Okay.

Ed

And then you really want to either try and avoid the clustering or try and use it to your advantage as best as possible.

Tom

Yeah. I was gonna say that as soon as you said that about, you know, it’s hard to break those clicks. It got me thinking, well, how, how can you, but I think the interesting there with, I guess, any form of targeting, you know, often the focus is on where to target, but just as valid as almost where not to target, you know, because there’s, there’s no point trying to penetrate something that’s not at the stage of, you know, that you’re able to kind of penetrate it.

Ed

And that’s about understanding your product as well. So when you’re in a product is really trying to replace a, you know, replace something you really want to build a community that supports you in doing that. Yeah. And then work on the edges or within that community. You know, if you think about it from like a sales point of view, you don’t wanna have to be fighting against the established product, every step, every sale, you don’t wanna be a, you know, a fight that you are the new guy that no one knows.

Tom

Yeah. Yeah. This is almost don’t, you know, don’t, don’t fight it, head on, you know, kind of

Ed

Exactly.

Tom

And I, cause I know you were talking about cliques in a, you know, in a very mathematical way as well. You know, a clique there is, it’s locked from the human side of them. You know, that that’s also representing a group. That’s probably got a very well, you know, entrenched opinion. That’s hard to break, but if I’m in a, in a clique within a social, certain social network, I’m gonna be exposed to many other social networks, you know, none of us exist in one network. So I guess the key then is making sure once those people, all the, all the members of that clique, if you like have been exposed enough times through their other networks, that’s the point at which that clique is probably open to, to kind of be penetrated if you like. So it’s almost not saying we can’t go after them now, but there’s almost a bit of planning, right? You, you, you’ve kind of gotta, you’ve gotta sort of follow where where’s available now and, and not waste resource where where’s, where they’re not available to, or they’re not open to your, your views.

Ed

Yes. Yeah. You’ve definitely gotta a yeah. A plan that yeah. And sort of monitor where that might be happening. And then also there’s another angle, I think, which is that you need to coordinate it, right. If you want to, you know, get your product into a clique, you have to try and get into the clique. You can’t think about selling to the individuals in that cl you have to be a coordinated effort to try and, and convince enough people at once almost. Yes. I mean, or alternatively you identify, try and identify people who think of themselves as a early adopters and things like this, you know, it has to be a coordinated effort into that community, as opposed to just hoping that what worked in sort of more disparate communities works there.

Tom

Yeah. And that, and that might be subdividing it. Right. Cause a cleat could be, um, let’s just, I don’t know, let’s take a technology like VMware, a random, random one. Right. But you know, back in the day when I used to work in that space, you know, they were very loyal community VMware, so trying to get them to move across would’ve, uh, on mass, would’ve been, you know, hard, but I, but I’m always thinking if you could isolate, there’s always subgroups, you know, you had the, the global community, the country communities, and then you had regional communities. So I guess almost how can you attack one small part of that larger CLE if you like, and do that multiple times over and, and, and break down the larger CLE that way,

Ed

That’s, that’s an example where kind of the clique’s working against you, but until the point it’s not.

Tom

Yeah, exactly.

Ed

And then it starts working for you.

Tom

Okay. That’s really, that’s, that’s really interesting. And I guess a question really do, do we need to change anything to do with our messaging, you know, to, to, when we talk to these different people or actually is, is, is kind of the messaging style less important, do we think,

Ed

I think the knowing what you are trying to do with your messaging, you know, you are cool to action. All the understanding that well is kind of is, is really key here. So are, are you motivating an extra sale or an extra purchase on the, on the consumer side? Or are you trying to replace something that already exists? Yeah. And in those cases, you know, your messaging is, is different. And then also like the role of other you, you know, sales and marketing tools you might use is also different. For example, like the role of a discount is very different, depending on where you sit, you know, in a copying scheme, you are kind of trying to convince someone to try something. So your discount is sort of like making that extra sale even more worth it for them. Right. And then you kind of have to, you know, you hope the, at your, the quality of your product means that they replace for complex copying where you have sort of a threshold star model.

Ed

Often there, we, we are talking about people who are kind of making larger purchases, and again, it’s sort of the discount might sort of motivate the purchase almost quicker. And that might also be the case in a survey system. Right? So the, in a survey, I think there’s, there’s sort of two uses for a discount either. It is to motivate like a newer update or, you know, that someone’s thinking about changing their behavior and it sort of motivates that to happen. But it also, because products in this space are often in a comparison, that’s where your discount is to be cheaper than the competitor, or not necessarily to be cheaper, but make the whole package cheaper. Right. A good example of this would be iPhones. I would say that iPhones at the moment are sort of making this transition into producing, you know, every now and then the iPhone basically comes out with a cheaper unit, like a, a lower spec phone and they do, they’re doing it again at the moment.

Ed

And it’s nearly, always when they’re losing market share. Right. And it’s because they understand that there’s a point at which there’s people who want iPhones, but the technology and Androids that you can get for your money E is enough. And I would, I would PO the iPhone, know that if the global share of iPhone fell below, probably not that lower level. So I don’t know what it is at the moment. It’s probably what, like nearly 50% in, in the UK, I would guess is about iPhone. It might, it might be a lot more than that. It might be a less, but I would say, you know, if iPhone went down to 10%, it might quite quickly go to zero because all the everyone starts conveying their friends. It’s like, oh, you’ve got an Android as well. Right. And then realizes that actually it’s not as clunky and hard to use as I most iPhone users seem to think it. So,

Tom

Uh, as a fascinating, for way to end on, I think that behavior spreading idea can also cause a product to collapse. I think that’s a really interesting point to, to finish on that’s it for this episode, almost a little bit of a, an experiment for this one, you know, I’m, I’m really intrigued from the podcast point of view for like our, you know, our new listeners, whether they found out about it from one friend or actually, did they hear it from a couple of people before they, they actually did it. So feel free to drop me an email. I’m Tom glab.ai. Um, and just let me know. We, do you think you fell into the simple or, or the complex, or actually did you survey all your friends about the, the podcast they listened to and, and was it a majority one to be you fascinated to know, but with your friends, you know, see, see if they need it to be a simple or a, a complex exchange for them to, to start listening and yeah, we’ll see you on the next one. Say goodbye, ed.

Ed

Goodbye.

Friends in conversation | Herdify

Sign up to the Herdify newsletter