Transcription
What? So, who's going to break first? I think it froze just, just as the countdown ended for me. Yeah, so I, that that threw me off. It's, it's good to be back. I mean, this is episode three in a row. Yeah, we're flying. I'm one. Well, you said we're flying. I'm wondering, is this going to feel a little bit like flogging a dead horse? Not given the topic that we're going to be doing, because it's exhilarating. It's exhilarating. It's super important. So, and of course, anyone clicking on this episode will already know what the topic is. So there's no need to give it too much of a buildup, because the title will reveal that we're going to be talking. Yeah, it's not this grand reveal. No, exactly. Yeah, we're going to be talking about segmentation, aren't we, Jamie? No, no, no fun anecdotes this time. Straight into it. Do you want to go straight into it? Yeah, I, I do actually, because we always say that we're going to be under half an hour and we never are. And I think it's because we're waffling on at the start of every episode. Yeah, I mean, we, we're recording this the day after, uh, Trump's inauguration. So we were talking about about that before this episode. But if we start to create funny anecdotes around Trump's inauguration, we're definitely not going to stick to the under half an hour. Yeah, true. And we'll probably alienate at least some listeners. Yeah, all three of them. We, we don't have the listeners to alienate, really. Like, unless, got to be careful, unless there's three listeners, one's left-leaning, one's center, and one's right-leaning, then at the very least, we lose one of them. Or maybe one and a half, depending on how, which, which we can't afford. We know. Let's keep the politics out of it then. Yeah, exactly. Maybe we'll have a little spin-off one time. No, that's such a terrible idea. We'll never do. Have a political spin-off of Sber emails. All right, let's, let's dive into it.
[Music]
Then, so segmentation. Yeah. Uh, and, and specifically, we want to talk about how to get started with segmentation. Because I think sometimes it's one of those things that people know about. I mean, you, you, you learn about segmentation as a, as a marketing concept, not just an email marketing concept, very early on if you study marketing. Um, but when, when people start a newsletter, for example, the last thing they're thinking about is segmentation, because they don't think it's for them just yet. And we want to talk about how you can get started and to make the little time that you have to invest worthwhile by picking out a couple of segments. Yeah, I think whenever we send newsletters, we are, we should imagine a newsletter is, you know, when they say panning for gold, yeah, the newsletter is the pan itself, right? It's not actually the gold. The gold comes from the panning. The newsletter is the pan. We shove it into the stream. I suppose in this metaphor, the stream is the audience, and out of that come chunklets of gold, which again, in this weird, convoluted metaphor, is data points or insight into our audience that we can then use. So I would say that when it comes to segmentation, view it as a means of panning for gold. And depending on when you start segmenting will depend on sort of how much gold you collect. And the hope is that as your audience grows, that stream gets bigger, you'll be able to collect more and more. Okay, let's recap on that metaphor. Yeah, so the, the, the pan is the newsletter. Yes. The gold is, is, is data points. Interesting data. Things we can learn about our audience. Okay. And the stream is the audience itself. Okay, fine. Cool. And, and segmentation is, is the, the process of panning for gold. Oh, so it's the, it's the motion. It's, it's the whole thing. Oh, it's the whole thing. Okay. Yeah, it's, it's the whole thing. Okay. So, so what, what I mean to say by that is that it's all well and good sending a newsletter, and that's, that's awesome. But if we really want to start using, getting the most value out of the newsletter, we should start segmenting. Okay. I feel like this metaphor starting to break down a bit. And the listener is the person panning for gold. Yes. And the stool that they sat on was panning for gold is the business infrastructure. I feel like, yeah, the weight of your interrogation is crumbling this this tower that I've built. This to for every metaphor from now on, we should test how flimsy is. Let's take it to extremes. Okay. So segmentation is, is a way in which we can, um, learn more about our audience. And by learning more about our audience, we can send them more relevant communications. And when we send them more relevant communications, it's better for the reader, uh, and it's better for the sender as well, because we're closer to achieving our goals. Yes, absolutely. And, and, and, and being able to send them more relevant communications is the tip of the iceberg. That's what we do within the four walls of email. But of course, with that data, we can also do things outside of email with, with that audience too. So if we learn that a certain demographic are interested in a particular topic in our emails, we might be able to target that demographic via our socials or other marketing channels. So, uh, there is a huge amount of value to be had in what it can do for our email marketing, but an extra layer of value on top of that for what it can do for our sort of wider marketing efforts. Yeah, I, I think one example of that is perhaps, uh, like retargeting ads. Yeah, totally. You know, um, so, and the, that is kind of, we are getting towards the more complicated end of of segmentation. But if we have data points on how our subscribers have behaved, then, you know, we can use that data to target them with specific ads outside of the female. Um, just like you described. But if we go back to how do we get started with segmentation, because that that kind of retargeting ads and you going outside the platform, that's kind of like the end goal. That's where we want to end up. That's how s. Okay, we want our segmentation to be eventually, right? And we'll by developing these quite, um, advanced profiles on our subscribers, then we know lots about them, traits, behaviors, etc. Um, but we need to start somewhere. So if we imagine, okay, I'm coming to you, James, and I have an email list. There are, let's say, for argument's sake, there's 2,000 people on my email list. And what I've been doing so far is the engagement metrics are good. And what I've been doing so far is emailing my whole list with my newsletter once a fortnight. Um, and now I'm starting to think, well, my aim to is to grow my newsletter audience even more this year. I'd like to double my audience this year. Um, and I think I might want to do some segmentation as a starting point. What would you advise from that point? Sure. So I think the big obstacle for someone in your hypothetical situation that you've just painted there, Jamie, is that they think to themselves, I don't actually have any data about this list. I think we need to take a step back for a second and explore the information we already have on those 2,000 in your list. Because ordinarily, we will definitely have their email address. We might have their first name. But depending on what information we're getting about them at the moment they sign up, that might be all we have about them. And on some signup forms, we'll have the company they work for, the amount they earn, the decision-making power they have within their business. There are all kinds of data points that you could get. Um, but most lists don't have those already. And that can put people off segmenting because they think, well, if I don't have this information, like the business they work for, how can I meaningfully, um, segment them? You know, how can I even begin to start using this data? And so, with that clarified, to answer your question, Jamie, what I would encourage anyone in your shoes to do is to use email engagement data as a means of segmentation instead of demographic information about the the the recipients. So what I mean by this is, look not at the sort of cold, hard facts of each individual subscriber, and look instead at the way in which they engage with your emails. And there are a couple of key metrics that I'd look at here. Firstly, I'd look at opens. And we've talked already about the ambiguity that can surround open rate. So they are to be taken with a pinch of salt to some extent. But we can create a, a segment, for example, of our most engaged recipients in terms of how frequently they open our emails. And that in and of itself can be interesting information. And I would also look at clicks. And we can get quite granular here. You can create a segment of everyone who's ever clicked on an email, or, you know, anyone who's clicked on one of your last five emails. But you could also, if you're getting really smart, and again, all of this information will be available to you via your the software you use, um, for sending emails, you will also be able to look at the specific things that people are clicking on. So if you've included a button in your email that says, click this button if you're interested in our services, and, you know, say seven people click it, well, that's seven presumably quite high-intent subscribers that could be segmented off. And I've intentionally used a low number here, Jamie, because we shouldn't worry too much about how, how many people constitute a worthwhile segment. If it's high-intent enough, it could be a really small number of people. Those seven people, we can send a follow-up email to and say, hey, I noticed that you clicked on this button. I haven't heard from you yet. Is there anything I can do to help? Or, you know, is there any information you need from me that might mean we can have a chat? Or, or, or whatever you, you, you want to put in that message. And so email engagement data can tell us a huge amount about our subscribers, even if all we have from them is an email address and a first name. I like that. And those seven people, if you're selling a high, high, very high ticket item, those seven people are incredibly valuable to you, right? If you're selling toilet roll, it's likely you'll have a bigger audience, and that, that, that number of people who have clicked are going to be higher. Um, and then you've automated it to try and get the purchase. But for those seven people, if you're selling a five, six, or seven, or even eight-figure thing, you know, those seven people are really high value to you. And that's the segment worth, worth looking at. Absolutely. And, and I really want to make it clear earlier in this episode, rather than at the end in our hot take section, which Jamie, you need to remember to do this episode. I want to make it really clear that we don't encourage segmentation for segmentation's sake. There's some advice that you'll read out there that says, every time you link to a blog post, for example, on your website, you should be segmenting anyone who clicks and then serving them with content that's similar. You know, there is an unlimited amount of granularity to segmentation. And if you really want, you can be, you could theoretically segment off the one person that clicks on your Twitter logo at the bottom of your email and send them a follow-up email saying, I noticed that you haven't followed me on Twitter yet. Um, at some point, we have to draw a line and ask ourselves the question, what is worthwhile from a time perspective? What is worthwhile segmenting and following up on, and what isn't? And I would argue that the 20 people who clicked on your blog post, unless the blog post is about buying from you, it's probably not worthwhile creating a separate email and going to the trouble and time of sending a separate email just for the purposes of of retargeting those 20. But as we've said, if it's seven people clicking on a really high-intent CTA, like, you know, use our services, um, that is definitely worth a follow-up in most instances. And so this, if you, if you're listening to this podcast thinking to yourself, my audience aren't big enough to segment, I would reframe that and say, instead, what segments can I create that are worth my time to follow up on? Nice. I like your point there about thinking, you know, drawing a line and not taking segmentation too far. And think it's almost, it's like that quote from, is it, I think it's Jurassic Park, where it's like, your scientists were so preoccupied with whether they could, they never asked themselves whether they should. Oh, that's, you just pulled that out from nowhere. I, so only because I, I love Jurassic Park. It's really weird. My, my girlfriend was watching Jurassic Park last night. Oh, really? Yeah. Which one? The first one. The very first one. Yeah. The one with the T-Rex on the toilet. Oh, yeah, yeah. Iconic scene. Yeah. Iconic scene. Um, not the Chris Pratt one. Not the new ones. No, I'm not that interested in those. They're good. Um, you should have been the one coming out with the Jurassic Park quotes then, not me. Yeah, I wasn't actually in the room. I was, I was practicing darts. Okay. Any, as I always do, I did hit 180. Okay. Yeah. Um, there's currently a 41 on the board behind me for anyone watching who might be wondering. There's a 41 on the board, which I think more accurately reflects my skill level as a dart player. Yeah, last time it was a 180, but it was a, I, you know, it was a manufactured 180. It wasn't a real one. This time, that was the last row I did was a 41, which I think is a lot a fairer representation. Yeah, unlucky, mate. Thanks. Um, okay, so we, we've spoken a little bit there already about segmentation and how to get started. So I think one thing that we need to highlight then is, if we want to segment, we need to start with data collection. Yes. And there, there are two ways in which which you can collect data. There's data that you can ask from people, and there's data that you can, um, determine from people and, and how they behave. So there are some examples of those. If we want to ask people for data, then that might be the signup form. Um, it might be in a follow-up campaign where we want to enrich our data, which typically has a very low uptake. Um, but it's where you're asking questions and people, your audience is is answering the questions, and then you know, you have the data there. The other kind of data collection we can do is by observing the way in which our subscribers behave. We then use that data to inform our email marketing efforts. So that might be like James described, engagement. It might be, um, it might be purchases. Um, it might be there certain things that they've clicked on. You know, there are things that people do. And actually, that's the much better for me. That's the higher quality and low effort, lower effort way to gather data. Yeah, I'm stealing this from one of our clients who described finding their ideal customer as like trying to find a needle in the haystack recently to us. And the issue that she was having is that, okay, she's got a relatively good list. They sell a very high ticket item, but the problem is that they are sending everything to everyone and finding it very hard to refine who's actually interested. And email data is one of the great indicators. It's the magnifying glass in the search because it gives us specific information about who exactly has clicked on particular links that show intent. And we can take that smaller demographic and refine our search accordingly and send them more intentional stuff, more, uh, specific stuff that really targets them and gets to the heart of why they've clicked but not gone on to convert. Nice. And I can see how that that is because you're, you're almost, you're reaching people in their natural habitat, in the way that they're behaving, you know, when they don't think they're being watched. And then you see people's kind of true intentions. And you know, if, if I sent out a questionnaire and on that, on that questionnaire, I, um, put a question like, do you see yourself using email marketing services in the next six months? And then, you know, the answer to that will be in my CRM. Um, it would be a much better indication of whether they're going to use email marketing efforts if they did or did not download a lead magnet that I'd written, a free resource that they then said, okay, well, actually, this is interesting, so I'll download it. My feeling is that people might say no to the first question, but then still download that lead magnet because they know that they're offering data that might subject them to a sales outreach message at some point. That's, that's a really good point. And you, you touched on it earlier, but I think it's just worth emphasizing the fact that whenever we're trying to gain data on our list retrospectively by asking them, it's incredibly difficult. So if we want to get demographic data on our list, let's say we've got, let's say you're listening to this and you've got 500 people on your list and you're thinking, bloody hell, I didn't ask them anything about themselves when they signed up. I'll send them a questionnaire. And, and this is a common approach that marketers used. I think I'll just send everyone a questionnaire. I'll probably get like 30% responses. The reality, if you manage to get 5% on a questionnaire response, you've done incredibly well. Because firstly, you have to depend on them opening the email in the first place. So already, you can, even in the best-case scenario, you're halving the amount of people you're actually reaching. And then they've got to take the time to fill out a questionnaire. Even if it's a minute long, that's a lot of friction. That's a lot of, that's a big ask of someone. So if you can get 5% to to fill out a questionnaire to give you more data, perhaps that's worthwhile for you. But I would argue back to your point, Jamie, that that data that they don't need to give to us, that we we can take from them, which kind of sounds more nefarious than I mean it to, that the data we can determine about them without their need for contribution is, uh, not only more valuable in a lot of cases and more honest, but also way easier. And because we can get more of it, it's, it's a lot more useful and a lot more valuable. Yeah, there are smarter ways as well to manually collect data beyond just sending a questionnaire. Um, you know, you might want to write up a resource and offer it for free. And the goal of that resource is to get new subscribers, but also to enrich the data of your current email list. And so it would be much better to say, look, I created this free one-page guide, you know, it's yours for free, you can have it if you want. And send that email to your email to your newsletter list. And if they click on the link, then they go to the website and they fill out a form to get that resource. And then you're enriching and updating their profile. So there are, there are more clever ways. And then you can talk about that one-page guide you've written for the next two or three months, you know, not just in one email as well. And so that's a much smarter way to manually collect data. Um, and by the way, couple it with growing your email list and getting new subscribers as well. And making sure the data that you collect, you're collecting all the data that you want from them at that point. And, and so this is my next question to you then, uh, James, is about, okay, we need to decide. If we decide all of a sudden, I want to do segmentation, right? I'm sat by the stream, I've got my pan, I'm ready to pan for gold. And I'm, that metaphor, I regret that. I was just thinking about that earlier. Like, that was such a mistake. I like it. I like it. Um, and so I'm ready to pan for gold, right? Um, but the signup form that I have at the moment is asking only for an email address. Would you advise at that point to say, okay, let's think about, okay, we can start to gather data based on behavior and observing people and how our subscribers behave, but we also want to address our signup form and how people are entering the list and make sure that anyone that comes in from this point onwards that we're gathering the data that that we want. Okay, I'm using a signup form as an example because it's the most obvious one. Signup forms, you have to be careful, right? Because you, you, you can't ask for too much because you're not giving enough away to justify asking for too much. So if we think about a signup form, what do we want to put on there? What's kind of a good port of call? What kind of data should we be asking for on a signup form? It really depends on your business. And if you are a B2B, a service-based business, for example, perhaps you're a marketing agency, the data that you want to ask at the point of signup will differ from the DTC brand, for example, or the, you know, the local pottery class where perhaps location or pottery skill is the most important metric to us. You know, if you're, if you're running a gym, then asking how regularly you exercise could be a key data point that you're going to use for for targeting further down the line. Um, so I think this is highly dependent on the business. One more objective answer that I can give though, Jamie, is that through our testing, we have learned that asking for three data points seems to be the the magic number. And that includes the email address itself, which is the most integral part. So if we ask for more than three, we start to see a decrease in the amount of people that bother to, like a meaningful decrease. There is a slight decrease between, um, one, two, and three, but it's, it's minor enough that it's well worth asking for the data in exchange for that that slight drop-off. After three, it starts to get more significant. And, you know, if we're asking for four data points or more, I don't know whether it's an effort-based issue or a privacy-based issue. I imagine it's probably a combination of the two, but there's something about four data points or more that that starts to put users off in a a noticeable and meaningful way. Okay, so yeah, I think part of that is because people might feel like it's slightly invasive and they start to get a little bit more cynical after data point number three. The other is that obviously, as you've already touched on, Jamie, it's a lot more effort to fill out this data for the user. And what they're getting in in in exchange is emails from a, a business, which isn't necessarily the most exciting prospect. Of course, they might be getting a discount off your services or a free resource or something like that. But, um, it's, it's rare that the incentive to sign up is so good that people are going to, are going to want to fill out more than than three data points. Yeah, uh, and, you know, if we take that, a signup, the example being signup forms, 100% agree. I'm going to give you an example in a sec about, um, a different way to gather data, um, that there's almost unlimited amount of data points you can, you can get from it. But, uh, just quickly, before I do that, if I'm selling, um, gold panning classes for beginners, what do I want to put on my signup form? Oh, you're never going to drop this, are you? I'm guessing, I'm guessing name, email, and, uh, level of skill or size of pan. Size of pan, that's a big one. Or the biggest nugget ever panned. Yeah, I, I don't, that wasn't a phrase I thought I'd ever going to say on this podcast. But biggest nugget ever panned. I, I would avoid asking people about their nuggets. I think that crosses the line. Right? Is that a privacy thing? Or like a, like a friction? I just don't want to talk about my nuggets. You know? Okay, fine. Um, the, the example that I alluded to there, where I think you can almost ask for an unlimited amount, within degree, within within reason, um, an unlimited amount of data points is, I filled out two questionnaires recently. And they're great. They're great initiatives. So if you are a marketing creative web dev agency, I would go and fill these out because what you get back is really valuable. One is the Wow Bench Press Survey. So the Wow company, they, it's a survey that, um, um, you, you fill out a survey questionnaire thing, and that then informs a report that gets sent around, and then you can see how you compare to all the other agencies that have filled it out. Uh, and Cactus also do a similar one, which I think is called something like State of Agency Nation or something. It's very, very similar. Um, and when I was filling out that questionnaire, I thought, Christ, the amount of data I'm giving both of these companies on me, on our operations, on revenue, on, you know, headcount, um, how I feel about the the market, you know, there must have been eight or nine pages full of questions. And, you know, so if we think, if we go beyond signup forms, there are ways in which you can gather data. I was, I was willing to fill that out because I, I wanted to get the report afterwards. Um, and I think it's a good initiative to, to, uh, contribute to. But when we think, so if you think about that, then all of a sudden, we, we're being a bit more clever than just doing a signup form. We're finding ways to gather data points. Um, I think one thing that I would say is that if we're asking for data, we should never ask for more than we actually need. Would you agree with that? Yeah, absolutely. I, I agree. I think that that's a really good raise, actually, Jamie. And there are, I think there are whole businesses that are dedicated to collecting data through that medium, and they're a lot less sophisticated than the examples you've just given. But it might be something like, if you're a Taylor Swift fan, for example, you might be on Facebook and you get served an ad that says, what sort of Taylor Swift fan are you? Take the quiz. And suddenly you're answering questions about, you know, how old you are and where, how many Taylor Swift concerts you've been to, and that kind of thing. And all of this is data that's being given away under the guise of this sort of fun quiz to find out what sort of Taylor Swift, you know, how, how much of a Taylor Swift fan you are. Um, but it, this stuff works, you know, and it's, and there's a, there's a natural curiosity that can be leveraged to some extent to to extrapolate more data out of people. I think, I think particularly the Bench Press report is a great example of that because people are ultimately quite curious to compare themselves to others. And that's, I think, a bit of sort of psychological marketing that that Bench Press does really, really well. Yeah, so, you know, I would just encourage people listening, you know, to think beyond signup forms and think about how can we gather data and in a different way. Um, I think we still have to set the expectation that, oh, by the way, as part of this, you're, you're signing up to our newsletter. But this is why our newsletter is so great, and it's kind of like an added bonus to, you know, I think because otherwise people won't expect to get your newsletter. Uh, it's not the worst thing in the world, but I think adding that in so that people know that as part of this, you're going to be emailing, emailing them. Um, I think helps. Yeah, and it, it's worth stressing that even, I think what I like about that so much, it, it is high effort for, for the, you know, that that a quiz probably isn't going to get you as many signups as just a quick email entry form. I think it's, it's worth stressing that, um, the reason why that sacrifice is worthwhile is because you get that extra data. But, but as you said, Jamie, you, you have to actually use that. You know, what we don't want to do is get fewer but more high-quality signups with with loads more data, but then just never use it. Because in that instance, we might as well just have the the quick email entry form where we're probably going to get more signups and a bigger list as a result. So, um, yeah, we, we absolutely need to use it. And I, I would say it's prudent for anyone who's really taking the prospect of segmentation seriously to start gathering this data sooner rather than later. Because even if now you've only got 50 people on your list and segmentation isn't worth your while, you know, in three years' time when you've got 5,000, you'll be really thankful to yourself that you invested some effort in collecting this data now and learning more about your audience now, because it will, it will reap dividends in the end. Okay, nice. Should we do hot takes? Let's do a hot take. Cool. Um, Jamie, go first. Are you worried about me stealing yours? Have you? No, no, no. I've, I've actually got one that we haven't touched on yet. So, okay. Um, yeah, so, so I'm, I'm confident you're not going to, you're not going to get it. Okay, fine. Fine. Okay, so my hot take then will be to, um, well, there's a few that I would like to choose actually. I think my hot take, let's choose one, will be start to observe behavior, um, of your subscribers and use that as a means to segment your audience. Nice. Okay. Okay. Mine is going to be start to think about how you can combine segmentation with automation to target audience demographics whilst you sleep. How does automation fit into the gold panning metaphor? Automation in the gold panning metaphor would be that you've, you've programmed some robots to do the panning for you. Wow. Okay. I like how this gold panning thing has turned into some like dystopian world. We need to do some like world-building and everything for this now. I think there's metap. Yeah, we're taking it far. We need to explore the logistics of like a waterproof robot as well. Yeah. Uh, there's a lot of things to consider there. I think Musk can help. Yeah, probably. God, the inauguration is fresh on his mind. Look, yeah, he's well. We can get Musk to help if he's, if he stops making mistakes like he did on the stage very recently. Yeah. Um, anyway, we said we wouldn't get political in this. We can all agree that it was a mistake. I think that's, uh, that's pretty apolitical to say that. Yeah, yeah, exactly. Um, okay, perfect. Anything else to talk about today? No, I don't think so. Um, no, I think we'll definitely do, um, I think this links very closely with personalization and automation. I think segmentation, when used with those two other big pillars of of email marketing, is even more powerful. So no doubt we'll do dedicated episodes about automations and personalization, but nice to be able to preempt them with this. Yeah, exactly. And there'll be more episodes to come about, you know, more specific segmentation tactics as well. We wanted to do one. I mean, just stay tuned for that. Bloody, I bet you I bet you're quivering with excitement over anticipation for those future episodes. But, um, I think, you know, this one, we wanted to make about how to get started. Um, and there'll be more advanced things to come down the line. But, uh, but yeah, should we go get some lunch? I'm starving. Yeah, me too. Thanks for listening everyone. Cheers.
[Music]