#285: Optimising Email Campaigns with AI and Machine Learning Insights

Welcome to Let’s Talk Loyalty, an Industry podcast for loyalty marketing professionals. I’m your host, Paula Thomas and if you work in loyalty marketing, join me every week to learn the latest ideas from loyalty specialists around the world.

This episode is sponsored by Epsilon. Today I’m delighted to announce a unique opportunity for one lucky listener of Let’s Talk Loyalty to enjoy a complimentary workshop with the loyalty experts at Epsilon. One brand every month will have the chance for a unique, independent loyalty lab. A review of your loyalty program where Epsilon will share their expert ideas, how to drive your program’s performance to a whole new level.

This workshop is a powerful way for you to measure and then increase the return on your investment in your loyalty program. So to apply, head over to letstalkloyalty.com/epsilon and enter your details.

Hello and welcome to episode 285 of Let’s Talk Loyalty, An episode dedicated to the importance of measuring and optimizing your email campaigns in order to further drive your customer loyalty. My guest has a background running loyalty programs for some of the best known retailers in Scandinavia, and through that he realized how much more could be achieved by leveraging both artificial intelligence and machine learning to understand the impact of every single variable in every single email campaign.

His approach is taking email analytics way beyond the basics of personalization, segmentation, and A/B testing that we all know and love. With digital marketing becoming increasingly expensive and first party data growing in importance, it really makes sense to find ways to leverage the latest analytical techniques to optimize your email newsletters.

Joining me to discuss the topic and share some of the latest drivers of success in email newsletters is Sondre Wassås, a partner at CPM Analytics in Oslo in Norway. I hope you enjoy our conversation.

Paula: So Sondre, welcome to Let’s Talk Loyalty.

Sondre: Thank you. Great to be here. Thank you.

Paula: Great. I know you’re a loyal listener of the show, so first of all, want to say thank you and acknowledge that.

Sondre: Yeah, I think I was part of the, like the third show or something like that. So since then I’ve been receiving your newsletters and, uh, listening to your show almost every week. So it’s been great.

Paula: Fantastic. Fantastic. And I know newsletters is the subject close to your heart, and specifically the reason that we’re meeting here today to share some, I think, very unique insights in terms of the impact of communications on driving customer loyalty. So before we get into all of this very interesting topic, let’s start with our usual opening question, and please do tell me what is your favorite loyalty program or even programs.

Sondre: Oh, you know, I’ve been part of so many programs, so over the years and, uh, it’s hard to choose from. But actually right now, if I would say my favorite today is, uh, actually Duolingo. I know. Do you know the sort of education app?

Paula: I do know Duolingo. Absolutely. Yes, Yes. It’s, uh, it’s for learning another language, isn’t it?

Sondre: Yeah, it is. So I know it comes in a free version as well, but eh, we bought a subscription from my family, so actually I have my wife and I have my kids in there as well. Uh, so we sort of have a friendly family competition. Okay. Yeah. And the thing I really like about it is they’re really good at motivating you to continue, to sort of train and exercise in a different language. And that’s sort of been the challenging part, at least for me when it comes to learning languages, is that you don’t practice enough. But I think that they are really, really good. So they have a lot of these lead boards, the tier levels that you have can go. And they also very good at reminding you that it’s only five minutes a day to improve your language skills and, nice. I think I’m on my day 85 now on my streak, so I don’t wanna miss, okay. Speak either. So it’s real good. So, uh, I’m not sure if I’m gonna be fluent in any languages, but it’s definitely gonna help me during vacations and things like that and understand a little bit of reading and then talk a little bit with locals and stuff like that. So that would be fun.

Paula: So tell us what language are you learning? Because I, I do love the fact that as a Norwegian, it’s almost like English is your own first language as well, because you guys speak it so well.

Sondre: Yeah, thank you. But yeah, the, the sort of the third language then that I’m trying to learn is, uh, Spanish. Yes. Uh, so, uh, I’m not gonna talk in Spanish here now, but, uh, No. Good. The thing is, uh, is just really fun. And that’s about it because I, I remember, you know, when you, you were younger and you were learning language at school. It was not the most interesting subject that you had in your school, right. So, but, uh, Duolingo, they’ve done a really good job of actually making it very interesting and just to keep, keep working on it and see small and small improvements in

Paula: Yeah. And, and you’ve reminded me honestly, yes, I have been very remiss myself. Um, you know, I do need to, uh, to learn and practice and really polish my French because, um, it is my husband’s first language and I need to be able to obviously speak to the family a lot more than having him translating everything back and forth.

So, you’re right, five minutes a day, we all can make that much time as a priority. So well done Duolingo. And I like that idea that you mentioned as well of it’s one thing to have gamification, but actually to have gamification within the family. I think that makes it, um, yeah, there, there there’s almost like this, um, this wonderful sense of, you know, it’s a shared experience.

So I think when we think about loyalty, there’s something in that which, uh, which is very powerful. So, uh, so that’s wonderful. Now, I think you mentioned you had a second favorite loyalty program as well, Sondre, didn’t you?

Sondre: Yeah, and I guess it’s kind of similar, but it’s called Zwift. So Zwift, that is like a training app.

Uh, so I like go biking. So actually you can put your bike, um, uh, on sort of a, like a spinner. Okay. And then it sort of helps you with gamification of your exercise. As well. Mm. Then you can sort of exercise against other people and you can do specific route and things like that. And they’re really good at sort of giving you health data and stuff like that back as well.

Okay. Uh, and that is really good for me because, uh, I sort of need the extra push when it comes to exercise. So whenever someone can give me a push, that’s really good. Okay. And but I’ll also say that things that I like about both these things is that this sort of reinforce or the why of why part of that program. Sure. So, It’s not about sending financial incentives or anything like that, it’s just okay, but I want to learn to read Spanish and they’re really good at sort of motivate me doing that. Yeah, and the same thing, I want to do exercise and they’re really good at motivated in me to do exercise as well, so that’s, the most important thing.

Paula: Fantastic. Yes. Well, as we know, it’s all about driving behavior change and in an ideal world it should be behavior change that benefits everybody. So definitely sounds like both of those apps are doing exactly that.

So listen, I know you are extremely passionate about loyalty. It’s been something you’ve been, um, working on since I think your university days. So maybe share with us first of all a little bit about your expertise, because you have done some incredible work, um, already in your career in terms of building programs and making sure to run them at scale across, I think all of the Nordics from what you’ve told me.

Um, so yeah, I’m very keen to understand that before we get into this whole topic about communication, so do share with the audience your whole kind of loyalty background.

Sondre: Yeah. So yeah, as I said, I started already on my master thesis where I was looking at the buyer, seller relationships and the length of them.

And uh, I was lucky. So I was, immediately went into sort of the e gaming business where you have a lot of customer data. Yeah. And in the e gaming business as well, you know that your competitor is only one click away. So it’s really important that you use the data you have, gaining the understanding of the needs of your customers. And then, as you say, try to communicate and do the behavioral change. Mm-hmm. . So I had the ability to do that, and I launched, uh, a few different, uh, loyalty program, uh, in the nors for Ladbrokes globally for, uh, poker brand. Actually, that was called Ladbrokes Poker. Okay. So, uh, and, but also in the retail, uh, industry, I’ve been part of, uh, Circle K. Uh, you might have heard like a large fuel and convenience retailer.

Paula: Of course. Yes. Yes. They’ve been a guest on the show and I know our, our mutual friend Mats Danielsen also you worked with, so I wanna give a shout out to Mats as well while, while we’re talking.

Sondre: Yeah, yeah. I’ve worked with, uh, Mats since many, many years. This is almost 10 years ago now, so time flies. But, uh, Indeed. Yeah. So I was part of the building that loyalty program, uh, from, from scratch, also being lucky enough to run it for quite a number of years as well. Mm. So we actually then the, the seven different markets in the Nordics, the Baltics, and Poland.

Yeah. So I think we were like four or five million, uh, active customers at the end. Mm-hmm. . Okay. So, and I also did the same for XXL, which is a large sports and outdoor retailer in the Nordex. Mm. So what I would say is that what fun thing is that I always sort of, I’ve been enthusiastic and I’ve been really motivated working in the intersection.

So between custom communication, analytics, um, and technology, and to me that sums of loyalty quite good. Because we work a lot with technology, we work with. Okay. Yeah. Uh, I need to have the data and understand the customer needs. Yeah. And get that insight for them to communicate with the customers to understand what drives their behavior and can affect their, their behavioral change for sure.

Um, so yeah, that’s been sort of my, that has been my passion for 20 years now. Uh, and I’ve been fantastic journey. Mm. But now I’m a little bit on the other side again, because now I want to focus much more on the analytical part. So Totally. There’s so much data out there. Yeah. Having a hard time sort utilizing all data.

But with the new tools and everything that’s available now, it’s so much fun that we can do with the customer data that we have.

Paula: Yeah, for sure. And I think what fascinat, fascinated me when you reached out Sondre, was, you know, this whole idea that I think there is so much awareness and appreciation of the power of particularly email marketing.

Um, but it seems that, perhaps you guys are one of the first to have this depth of analytics into, you know, the actual wording, the layouts, the formatting. So some of the, I suppose, expertise around email marketing is very well established, but I feel like you guys are taking it to the whole next level. So will you talk us through, first of all, Why are you focusing on email?

I sometimes think that brands tend to think it’s, you know, email marketing has been around for 20 years now and maybe it’s been taken over by social media or you know, lots of other forms of communication. So why is it that you’re focusing on email particularly?

Sondre: Well, I feel like emails that it comes, goes in waves and, and it has this lower, high and a little bit over lows, but right now it’s becoming very, very important again for companies.

So we see for our point of view, a very high demand on understanding more on the effect of emails. And it comes from, I think most of us agree that it’s the most profitable marketing channel that you do have. So if you’re looking at sort of profitability and sort of the environment that we are in and going towards now, which we know is gonna be tougher for a lot of companies, you sort of, you need to sort of squeeze the last thing out of the lemon Yeah. To sure that you stay profitable. Yeah, At same time, what you also experienced is that all digital marketing are becoming much more expensive. So if you look at the Google, Facebook, Snapchat, TikTok, et cetera, Yeah. They’re also increasing their prices. Mm-hmm. . So when you combine these too, we see a lot of effort now going back into email marketing, making sure that they can sort of generate the most effect out of their existing customers as possible.

Yeah, so that’s why what we actually, we are a company called now, CPM Analytics. So CPM stands for Customer Portfolio Management. So we work with customers and we work with analytics and we have been working a lot with sort of text and image analysis for a lot of our customers. Lately we are sort of focusing a lot about, uh, the email marking as well, because I think that we all, we all have the feeling and we think that it’s important to have a good text and a good copy, and it’s important that the design is good, but what is the effect of having a good copy and a good assignment?

That’s something that I don’t think anyone sort of have been able to analyze and, um, give us answers to previously. So what we now are doing is that for our clients, and we, we have numerous clients here in the Nordics now, that what they’re doing is that they’re sending us their. Last 2, 3, 4 years of email data.

Mm-hmm. . And we are doing then a full text image analysis of all the data they have previously done. And we find some very, very interesting information about how the different emails actually work and what doesn’t work when it comes to, when it comes to sort of the texts they put in. And it’s, I think as well when we present the results, um, most of them sort of get a little bit blown away, but the same time they set us, but we, we had it somehow in a gut feel that this was a challenge that we had, but we never have been able to sort of put into any analytics. So, uh, I think it’s great product now, and we see that there’s a lot of, a lot of organizations out there looking for it. So yeah, it’s, it’s within retail, it’s within. University has been in hospitals, It’s everyone is sort of asking for, for this type of analysis, which is fine.

Paula: Yeah, yeah. No, it’s good to hear that. You know, I suppose the, the gut instinct is there you are building on that by, I suppose, delivering ideas and insights that do seem to be blowing their minds. And we’re gonna talk through some of those. But even just to go back to your point as well, Sondre, about, you know, the market size for email is, is only continuing to grow.

You shared with me, it’s what it’s, it’s 15% annual growth rate, still compounding. So were you surprised that that is still continuing to grow at that scale, given how mature it is? As as a marketing tool?

Sondre: Yes. But also at the same time, no, because now we are getting all these machines that helps up with sort of the automations, the personalization, the segmentations.

Yeah. And we’re starting to be even better at proving the value of the newsletter. So I think that the value had probably always been there. Yeah. But sometimes it got forgotten by Google came with their sort of analytical tool and we started looking at all the social media and what they were really good at was they were good at showing the value. Mm, having your ads, that platforms, but now that you have the email tools as well. Yeah. Which is more sophisticated in this area. And when you now start to compare, we see when we do our analysis, that email can be up to 40 times, 40 times more profitable than social media. My God. And that’s massive when you have a lot of companies now sort of searching for any revenue, any business that they can get.

Yeah. Well, of course, it’s 40 times more profitable than just making small adjustments. To make a huge sense. So yeah, it’s definitely on a high right now. I think also the numbers were it’s 15% annual growth rate, but they were also estimating that will soon become $20 billion industry.

Paula: My goodness. My goodness. And I think you’re right. I do think there has been some things that are very well understood that, um, you know, I guess because it is relatively mature, I think we all know that our clickthrough rates and our open rates, are important metrics and I do think we’ve, we’ve all done plenty of work on personalization and the automation and the testing.

So will you talk us through, I suppose the parts that you are focused on in terms of your analytical piece that you don’t think has been researched to this extent before, that’s driving these mind blowing insights for your clients?

Sondre: Yeah, so what we are looking at, pretty much if you’re looking at a newsletter or an email, it’s three key components that you do have.

Mm-hmm. . So you sort of have the messaging and the content. Mm-hmm. , which very often you will then say is text. Yeah. And you have the sign. The sign sort of how you are using images that you put into the email. Mm-hmm. and then what we call them functional variables. That sort of when do I want to send it to whom are, do I want to send the email?

So if you look at these three, we can then go into looking at further into the messaging and content and then we start deciphering and looking at, okay, but what is the topic of the newsletter? And then this machine learning. Mm-hmm. . Finding out the topic of the, uh, of the email. Mm-hmm. other specific words in the subject lines that sort of trigger words.

Mm-hmm. that sort of have any effect, uh, on it. And, um, we can do all different kinds of things, uh, as well. And looking at the variation of the contact, how complex is it to really, you know, someone writes very simple emails as almost like a child book. Okay. Others? Do much more complicated emails, which is more, uh, like a textbook.

Mm-hmm. . What that matter and what is great with these things is that it’s almost always a variation, of how you write. So you don’t write the same thing every week. Mm-hmm. . So how you write will differ, and for all of us doing analytics, we love variation. That’s sort of what we find results from. So, um, I can go through a few of the examples of findings later, but a lot of interesting things there.

So if that’s sort of the content messaging part, We also look at the sign part, and there we also use, um, artificial intelligence as well. Mm. Where, And the machines actually then start to read the images. Mm-hmm. . So it actually says, okay, here we have a face. Here, we don’t have a face. Here is a happy expression or a neutral expression.

Yeah. I can see that’s, uh, it’s uh, you can see the age of the person. We can look at the gender of the person, person, et cetera, and in what kind of environment the person’s in as well. And then we start to figure out does that matter when it comes to the effect of the email? And that’s super interesting.

Yeah. Finding out that, Okay, if you have actually findings that we have for a client was that female generally performed better. In an image, than men. So pictures of women. Yeah. Yeah.

It doesn’t matter if it was men or women that opened an email. Okay. Still from better. Interesting.

And that’s super interesting for them to know because yeah, without that information, it might be coincidence what type of images that you’re using.

But now we can say, actually you should have more women. They should have a happy expression, not a neutral one. Okay. cause that also have an effect. And what I also would say is that we, we call it effect analysis. And that’s because we look at all these variables at the same time. So to find if it’s significant findings that you have.

Yeah. So an example that just gonna, that sort of very easy to understand is that we found that, uh, the word, um, In a week, as in a weekday. Mm-hmm. to be very promising trigger word. Okay. But then when we look at all the other variables, we found that, okay. But that word happened very often at the end of November.

Paula: Okay.

Sondre: And you know why? Mm. No tell me. Black week. Okay. Coming up to Yes, Yes, of course. . So

emails sent during Black Week perform better. So that’s because of season. yeah, not because of the word week. Okay. So, uh, and that’s what we learned, show them as well, is that it’s not, Uh, sort of only bivariate analysis where you look at sort of one factor against another.

We look at all the different factors at the same time to find out okay, which one is significant. So when we come back to the customers, yeah, with, uh, great deal of confidence. Tell them you need to do more of this and less of that. And that’s very interesting.

Paula: For sure. Well, you’ve definitely got a bigger analytical brain than me, Sondre.

I mean, I, I love the, the insights, the fact obviously that you’re using the latest technology, artificial intelligence and machine learning. And again, I suppose with my own much more limited expertise, let’s say an experience we would probably only have done like A/B testing. So again, we would’ve understood that there was important differences.

But the amount of time it would take to test every single variable in an email, you know, whether it’s from the subject line to, you know, the, the style of image, you know, there’s just no way that you could actually deliver without the benefit of these technologies. So it sounds like. It’s something that you’re able to do all in one go. If you look back, as you said, I mean that’s quite incredible to look back at a portfolio of email communications over a couple of years. And am I right in understanding then that you’re coming up with a list of recommendations for those brands to basically say, okay, your trigger words, are these, the style of imagery you need to use, are these, so it’s almost like a playbook for their email marketing is it?

Sondre: Exactly. I think it was very well summarized.

Paula: Good. Yeah, I’m glad I followed.

Sondre: Well, I must say as well is that we, we really sort of encourage, uh, testing and optimization. So everyone needs to continue to do that. So that’s something that’s very important for anyone doing email marketing. Um, but, what we do is exactly what you say.

Even without having a testing strategy, we can take all of your existing emails and actually find the insight from all of that. Because as I said earlier, there are natural variants in messaging, content in design. Yeah. And when you send the email, so, and what we then come with them is like a playbook. So, uh, good word for it, because, It shows them, um, examples like there are specific topics that when you write about them mm-hmm. has a better impact or a worse impact than other topics. And then what’s interesting to do then is that, uh, what we, what we usually say is that when you find one answer, You get more questions, ,

Paula: You know, that’s, that’s actually sometimes cause us more confusion.

So I hope you have a solution for that as well.

Sondre: No, that’s, I think that’s actually good because it creates awareness. So if you find out a topic that’s not doing well. Yeah. Then you need to know why doesn’t this topic do well? Yeah. Is it how we communicate, Is it how we using images, Is it when we sending the emails or is it, uh, actually a topic that’s not out of the interest for, uh, our customers?

Mm. Or perhaps it’s the interest of a very, very small niche of our customer database. So we need to segment these type of topics to a greater extent. Yeah. And that’s also how you are sort of learning how you’re becoming better at doing email marketing. I usually, in my, in my career, I’ve been working with email marketing since 2004, so almost 20 years.

Yeah. And so I had one thing that said that, um, that I’ve done and to my team is said that we need to test everything because it’s all about learning. Uh, said, um, I don’t have a silver bullet to tell you. This is how we’re gonna be extremely much more profitable. Yeah. On versus just gradually every day become a little bit better.

Yeah. But imagine now actually what I can go back to customers and say is that I can within a few weeks, generate three years of learning. Yeah, that you can get, and then you can sort of put into your communications to increase effect of your marketing. It’s, yeah, it is fantastic.

Paula: Totally. So, so the two that I’ve heard you mention so far, and again, hopefully I’ve got these correct.

First of all, it seems conclusive that when you’re using imagery that, uh, pictures of women, perform better. Pictures of smiling women perform even better and definitely no neutral faces. So that seems to be important in terms of all of your analysis. Is that kind of a, a good conclusion on the imagery side?

Sondre: Well, I think you have to be very careful of generalization. Okay. . So, uh, here’s sort of a heads up and that’s if you sort of try to Google best practices. Yeah, be very careful of what you’re reading, because it depends. Different clients of us mm-hmm. get different results. Okay. So we are not coming up with sort of an industry standard of this is exactly how we are gonna do.

The example that I had was for a client of us mm-hmm. was definite for them. It was significant. Okay. I think that’s important as well because there’s a lot of sort of variation between industries within industries, within different brands, etcetera. Yep. And that’s why, and that’s actually the reason why we started doing this because you could read, um, if you could read sort of advice online saying you need to have a large image on your email. You need to use this type of text. And we were sort of asking ourselves questions about, is this correct? Is, is this right? And what we now see is that, Um, No, you can’t say that this is how you’re gonna write your emails because it depends.

You need to look at yourself, your brand, your customers, as an example. Okay. And I can take, um, an example is that we, I can use one of our clients, um, which I used to work for, so this is use and which is, uh, XXL.

Paula: Okay. That’s sport retailer. Yeah?

Sondre: Most retailers, if you look at their emails, they are almost no content in text in them. It’s images of products with prices.

And that works extremely well for them because their consumers, what they want is sort of, um, a summary of the best offers for that week. Okay? But within the same industry, sports industry still, you have others that see the opposite effect, that actually some of the consumers are interested in, in reading about products. Um, rather than just seeing, uh, product and prices.

Interesting. So, okay. You will have differences. Uh, but, uh, I guess the, the women and smiling is probably one of the more generic ones.

Paula: Okay. Okay. Well, it makes sense to me. Uh, definitely it feels like the kind of thing that I would respond to. So what else have you noticed then, like, again, I know we’re, we’re saying we can’t generalize, but you have a lot of findings across all of this, um, expertise.

So what else, um, interested you, you’ve already talked about the word week. Um, and not to assume that that was a trigger word, but, I do think it’s very interesting that there are trigger words as distinct from, you know, just, you know, a, a subject line, for example, which I, I guess, is very easy to test. Um, I think we, we’ve probably all done that, but trigger words is something that I guess probably does need the machine learning in terms of understanding and analyzing it, because I’ve never known trigger words to be something that would be easily measurable, certainly in anything that I’ve worked on.

Sondre: No, and that’s right as well. So, um, what I can also say is that a very important word that I think most within retail actually think about is that, should I use save, should I use discount, uh, or are there other kind of similar words that I should be using in my subject line. What works. Mm-hmm. . Yeah. Uh, and, uh, this was done for a specific client this time It was in Norway.

Okay. But they, what we saw was a significant fund that using the word save, and then the amount had a negative correlation, on, um, the effect of the news. So they should do others, use discount, um, use the percentages, the amount etc. rather than using the word save because for some reason, um, uh, that doesn’t, um, do well with their customers in Norway.

So that’s the kind of interesting word. So, try to avoid using that one and find someone else as well. Uh, instead. But on the subject lines as well, just the sort of continuous, sort of interesting information is that, uh, we also saw, um, quite a few actually of our clients have what we call low variation in subject lines.

So what we then do is have, we look at all the subject lines that have had for the last 30 days, we see mm-hmm. . Have they had a similar subject line in the previous 30 days? Okay. And it doesn’t have to be the same because you know, the machine learning, it understands the intention of the sentence. Okay. So even though you use kind of different words, it’s, um, it still say about the intention of the, uh, subject line is still the same, so it’s small variation, but then the problem we then saw is that low variation, gives lower opening rates. And so, of the reason for that is obviously that the consumer and the reader, mm-hmm. feels that he has seen this message before.

So, you know, in your cluttered email box Yeah. You scroll through very quickly all the emails that you get. Yeah. But I read this before, so you don’t have to remember the worst of in there, but sort of, Yeah. I got this type of message from this, um, from, uh, this, um, company before. Okay. We don’t open it. So actually what we saw was, uh, sort of in, uh, in relative numbers, it was 10% higher opening rate when they use subject line with varition.

So, Okay. Many people might think that it’s not that important to write the perfect subject line, is it? I can just do kind of similar what I did last week. Yeah. Maybe findings like, Ooh, you should definitely put some effort into your subject lines because that will have a huge impact. Uh, because you then look at the full funnel as well.

If you don’t open, you don’t get a click, you don’t get a visit to your website, you don’t get that sale, we all work for.

Paula: And, and what strikes me with that, actually it reminds me of, um, somebody on LinkedIn who, um, I really do admire in terms of their content. But it was, um, the same point again, just in a different context.

Um, this was particularly, I suppose, somebody, you know, I probably would kind of consider them an influencer, let’s say. And what she was talking about was the importance of a surprising image. So I think particularly on LinkedIn, you know, we all see a lot of the same stuff and we see a lot of text and there’s, I suppose, almost a formula now.

But I think what I’m hearing with, I, I, I hear him hearing you describing it, is variation, but. , it’s almost that there needs to be an element of surprise to cut through for them to go, Oh, that looks interesting. What are they talking about? That’s, that’s very different rather than, because I do think everybody does realize the subject line is important.

Um, so they definitely, I think we all put kind of work into it, but, but writing a surprising subject line is not something that I’ve ever sat down to try and do. Um, so yeah. Is it, am I right in understanding that one?

Sondre: It’s all about catching the attention of the reader, uh, and that, uh, and then I think having sort of a surprise, um, yeah, definitely help you.

Um, we also say that you have approximately two seconds of getting the interest of the reader in the newsletter. So yeah, you just think about two seconds. What should I write if someone has two seconds.

Paula: Totally. Yeah.

Sondre: Um, then perhaps it makes a little bit more sense that sometimes if you think that you ready before you just scroll on, you don’t open it.

Uh, if it catch your eye and say, What is this, this, this sounds interesting. Yeah. Then you engage.

Paula: Totally. You’re making me think about my podcast descriptions. Now, Sondre, I don’t know how, uh, how surprising I can make them three times a week, but it’s certainly something I’m gonna be much more aware of.

Sondre: Yeah, just, uh, I sent you as well, and uh, I sent you the slide deck and I put the first email that I got from you. Yeah, The last one. Yes, indeed. For example. See, does it differ? Yes. Or quite similar.

Paula: Yes, exactly. Yeah. Well, I would definitely say from our perspective, we have extremely limited resources, given that I’m pretty much doing almost everything, thankfully now with, uh, with two guys in the team supporting me.

But yes, we, uh, we’re, we’re covering a lot. So, um, we haven’t got into the depth of analysis of, uh, writing surprising email subject lines or podcast descriptions. But, uh, I love to learn about it. And then the other piece that you talked about a lot is the simplicity of the content. And I guess for me, you know, when I think about how copywriters typically work, particularly in bigger companies, and you’ve talked about XXL, you’ve talked about Circle K, so I’m sure these are brands that do have, you know, brand guidelines that are very well established. So in my experience, the copywriters and companies like this, you know, it’s probably a consistent style of copywriting that we would tend to use maybe across all channels. But it seems important from your work that actually it is simplified, uh, in terms of the email needs to be a little bit simpler if it’s going to have a better effect.

Is, is that, am I understanding that point correctly?

Sondre: Yeah, I, I, Well, I think it’s least what you need to do is, so you need to write at the level your customers are at. And if you don’t think about what I said earlier, is that they, you know, that your customers and your consumers have limited time to spend.

So it needs to sort of be quick and understandable immediately. Okay. Um, but it’s not as easy to say that simpler is better either. Uh, because if you go all the way down to very simple language where sort say, Okay, I’m max gonna have five words in a sentence, and almost like a child book written. And then you are not gonna hit the right level because then the consumers will write it.

Now they read it, uh, and think it’s not for me. Mm. So, uh, within there, actually it’s, we see significant findings there as well. So you need to sort of find your level. So, but it is more, I would say it is not child books, but it is more on, on, um, magazine level. Okay. So we tend to say that there are four different levels.

You have child book, you have magazine. You have newspaper, Okay. And then you have sort of more professional writing. And then keep it sort of on a magazine level, uh, that seems to be where most effective. So sometimes you write too simplistic, sometimes you read way too complex. Yeah. Find your right level of writing.

Paula: Yeah. And yeah, of course, the point is it depends on the sector of course, as well. And as you’ve already, you know, mentioned the, the particular brand, but you know, I’m a B2B marketer, so I’m certainly not going to write too simplistically, um, or even probably magazine style because that to me feels like a, a consumer style of writing.

But I know a lot of your clients are in the retail sector and again, your own career. In, in that. So that magazine style makes perfect sense, and I like that classification of the four different levels. So again, I need to sit down and think about my, my emails a lot more. Um, so much going on. So listen, um, there’s obviously a lot of, uh, opportunity for, for people who are listening to really optimize their email performance.

Um, as you said at the very beginning, Sondre. It’s absolutely this case where it’s the single most profitable opportunity I think that most brands have available. It almost sounds like, you know, once this channel has been explored and let’s say optimized, I’m pretty sure you guys are gonna be applying similar expertise across other communication channels.

Uh, who knows whether it’s social media or podcast. So who knows what in the future. So, um, super excited to hear that. I suppose the role of communications in loyalty is getting this level of focus and expertise, um, because you are the, certainly the first company and the first agency that I’ve met that has really focused to this degree with an analytical skill set.

Remind me how many analysts you have working there with you in the company. Sa cuz I know it’s a big team of, of um, very specific skills on the analytics side.

Sondre: Yeah. I mean they are getting analytics these days and getting people like with the right background, it is quite challenging, but we have, of the senior team, it’s 10.

Amazing, good analysts. And it’s not about just doing the analytical part, it’s data handling, data monitoring. Yeah. That part as well, uh, to make this as effective as possible. So, um, is is a really great team. You know, I’ve, I just a short story on that in my previous, teams that I’ve been managing as well, I’ve been responsible for trying to also build up the insight teams, both for XXL and of the teams as well, and, I never understood how far I was from succeeding until, until I came here, because I always thought like, Okay, okay, we are a few people here now we’re doing analytics. We get a few new reports. Now I want to move over more into data science and understand more of predictability. If I can sort of predict what’s gonna happen and I want to do more social science, understand why is this happening and why will continue to be like this.

Mm-hmm. . And so I never bridged that gap. Uh, but that’s because I never understood what it took until I came here and sort of see what these people are, are able to do.

Paula: Yeah, and I think as well, your particular skill seems to be the storytelling side. So translating the, you know, big data analytics, um, all of these implications.

Into something that’s actually actionable. Because for me, research can sometimes end up leaving me with, Okay, so now what do I do? Like, so, so, so what I really think is super important is to have that storytelling piece as well. So I know that’s something that you pride yourself on, on something, that you’re able to translate it into something that’s usable.

Sondre: And I think that’s something that everyone that works with loyalty has to master because loyalty is such a broad area of expertise that goes loyalty marketing. Yeah. And you need to do the storytelling, and uh, I learned that through, through my 20 years is that when you need to have big organizations, you have hundreds and thousands of people that needs to walk in one direction. Mm-hmm. , you need to be able to deliver a very strong, good story that everyone believes in. For sure. For it to be successful. So, uh, yeah, can’t underestimate the, the sort of, the value of storytelling.

Paula: For sure. For sure. So listen, Sondre, that’s the, um, that’s all of the questions that I have for you today. Um, was there anything else that you think is important for our audience that you want to mention before we wrap up?

Sondre: Uh, I think I’m really happy that you’re so quickly sort of understood, so the thing that we’re trying to do, the value that we bring, and also that you mentioned that through the communication channels, that now we’re doing emails, what we can do, social media and stuff like that as well. Because what we actually are looking into is what is the effect of messaging content?

What is the effect of the image that you’re using? Yeah. And think about that as well. The machines. They tell you what type of background, color and, and the color in front you should have on an image. So it might say that you should have an image of someone in white on a brown background that has a greater effect.

Wow. And imagine then applying that to as suicide social media positive. It’s, uh, Instagram, if it’s Facebook, if it’s Snapchat, wherever it is. Yeah. Uh, we all are affected by different things, uh, but they can be the same for many people, but we don’t know what they are until we can analyze it. Mm. So now we can.

Amazing. Great. And a little bit scary at the same time.

Paula: I was gonna say you’re smiling. I can see that you understand and enjoy the complexity of it. Um, so again, I love, uh, working with people like you who can translate things into, uh, into my language, into marketing speak into consumer implications and insights as well that I can actually take action on.

So I, I just wanna thank you, um, for, for joining us, for reaching out to me. Um, I will of course, make sure that we link your profile in the show notes for this particular episode so people can connect with you there in Norway. So with that said, I want to say Sondre Wassås partner at CPM Analytics. Thank you so much from Let’s Talk loyalty.

Sondre: Thank you. Thank you so much having me. It was great fun.

This show is sponsored by the Wise Marketer, the world’s most popular source of loyalty marketing news, insights, and research. The Wise Marketer also offers loyalty, marketing training, through its Loyalty Academy, which has already certified over 245 executives in 27 countries as certified loyalty marketing professionals.

For more information, check out thewisemarketer.com and loyaltyacademy.org.

Thank you so much for listening to this episode of Let’s Talk Loyalty. If you’d like us to send you the latest shows each week, simply sign up for the Let’s Talk Loyalty Newsletter on letstalkloyalty.com. And we’ll send our best episodes straight to your inbox, and don’t forget that you can follow Let’s Talk Loyalty on any of your favorite podcast platforms, and of course we’d love for you to share your feedback and reviews.

Thanks again for supporting the show.