Revlifter
So first of all, thank you to everybody that's joined today. Today I'm gonna be talking about achieving scalable deals, personalisation in affiliate marketing, and I'm gonna do that from an affiliate angle. But I also wanna keep in mind that, of course, everything I'm talking about really is affluent across affiliate marketing in general, across eCommerce in general. So what we're gonna talk about today, I just really wanna first and we'll define what we believe personalisation is against what actually I think the market probably thinks it is. Then I'm gonna talk a bit about customer expectations. So what's the data value exchange and we're living in a post GDPR world now. I wanna talk a bit about what this new world order really looks like, and then also personalisation in practice. How do we actually practically understand how advertisers are currently using personalisation? Then also just case studies that we've got, just to talk about how we believe that personalisation is developing and how we are pushing the boundaries within that sector. Then I'm gonna talk a bit about what success formula looks like, we need to understand also have improving incrementality. And then I'm gonna talk a bit about where I believe the future of personalisation is actually heading. So to begin with then when we talk about personalisation, Google says at least that the action of designing or producing something to meet someone's individual requirements. When we look at personalisation from that angle, it actually talks about it being an individualistic thing. Actually we believe there are two parts to this equation. So what we say at Revlifter is personalisation is the process of creating one-to-one experiences for customers, presenting consumers with unique deals, which are specifically tailored to their needs and desires all based on the retailers goals. What do I mean by that? What I'm saying is actually we believe that personalisation has to start with the advertiser, cause we need to understand what your objectives are, Initially, cause we can't really start to actually then present personalised offers to the consumer until we know what you are actually trying to achieve. And what I mean by that is understanding things like not giving discounts unless we know what the margin is and the product, not being detrimental to the brand equity for example, and then understanding also how do we change consumer behavior to benefit you, the advertiser, as well as also what the consumer ultimately wants, which is the personalised deal. So we think about it very differently, we don't believe it's a one way street, it has to be a two way street. We believe that there is a real shift in customer expectation. Consumers understand that there is a data, there's a value exchange of their data. They're willing to use it as a commodity in exchange for personalised deals. I'll explain how that actually works in a moment. So to give you some ideas around when we talk about clear value exchange, if you look at brands like Clinique, for example, where they're asking very specific questions to their audiences around the types of skin tone you have, what type of makeup you're actually looking for and actually ascertaining what type of foundation ultimately is right for you. Coming back with the ability to once there's been a clear value exchange and the consumer's given up some of their personal preferences, they're able to then to retarget and remarket to those consumers. Now, not all advertisers are doing it, but I would say that's the benchmark. It's the bare minimum that we should start thinking about doing within the affiliate channel and consumers are expecting to be asked those types of questions, but then on top of that as well, actually a brand that I recently discovered through this journey of understanding how personalisation really works is Thread. Thread actually ask a series of questions at least 20 to 30 before you can even shop. Now, I don't know any brands really that are able to actually ascertain that much information before consumers are really willing to give up and shop. But the experience for me was I looked at initiative ideas of types of clothing that I'd like to buy. And at the end of it, you actually get allocated a personal shopper, a personal assistant who actually is a stylist for you. Even though it's an automated email that gets thrown out, you're actually able to contact them directly. With everything there needs to be some kind of threshold or spend. So for example, here saying you spend 50 pounds, you'll get a 15 pound gift card. And ultimately what we're doing is we're also setting a threshold of an average order value for the type of customer that we're trying to actually capture as well. However, I think the market has moved on significantly from that, but when we talk to our customers here at Revlifter, we actually ask the retailers why they wanted personalisation. Now, unsurprisingly of course, most of them said are more average order values, increasing conversion rates. And that's what you would expect, especially from the affiliate channel being sort of towards the bottom of the funnel in the sense it's a conversion channel, however, what we started to actually get from advertisers when we delved a little bit deeper was they wanted brand loyalty. They want customers to come back and shop with them. They want to really understand who their customers are and actually have a rapport with them, actually have a dialogue and I think the industry's really moving on, because we have a significant amount of data, which if you go back five years ago, we really struggled to process. Especially when I was client side, one thing that I really struggled with were the different systems that we had in house. We really struggled with the amount of data that we were capturing, what to do with that data. We're in a position now where we can just skim the top of it and really understand and ascertain what type of consumers actually bringing to our sites. So, how do we actually go about aligning to the retailer's goals? Initially, here at Revlifter, we believe that we really need to understand exactly what people are putting in their baskets, not just trying to stop them leaving our sites, which is what we've really been worried about for the last 10 years. And that's where technology has been developing, but customers don't necessarily just want to leave the site and go somewhere else. They're just looking for a good deal. So what we do here is we understand what's in the basket. These retailer data signals allow us to actually then start to capture and market to our consumers in a very unique way, because if we understand what our basket thresholds are and what people are putting in their actual baskets themselves, we can actually start to skim that data layer and put it into a feed engine. Most of our partners will actually scrape a page, we don't believe that's the right way to go. And what we then need to do is we need to execute on the offers using our if and am statements, but ultimately machine learning, which is our buzzword at the moment. At Revlifter, we believe that actually machine learning is all about understanding the consumer behaviour onsite, but also executing all the types of deals that ascertains to what's in their basket. So I'm gonna talk a bit about some practical examples in the industry of where I believe there are good, solid industry benchmark, but it's the bare minimum that we should be doing. I wanna start really from the consumer point of view, before we slip onto the advertiser point of view. So a really good example of understanding how we segment our customers is Hotels.com. The Hotels.com rewards program allows customers to have different tiers or different levels based upon how many nights they stay, so for every 10 that you stay, you receive one hotel night. Now, what this allows Hotels.com to do is ultimately market to their customers with very particular deals cause they know what hotels they're staying in, which is a program that's now been copied throughout the travel industry. However, as a very bare minimum, though, I would say that what hasn't really developed is the ability to target them at a tiered level and actually start to push people into gold and platinum memberships, and we really struggle to manage that because it's very difficult to market to consumers within an industry and make them buy more. If you can't understand if they're on a business trip, or if they're on a personal leisure trip. So if you look Booking.com, they're the masters of persuasive nudging, one thing that they do particularly well is understanding the occupancy rates of their hotels. Now, this really is pushing consumers to actually book because the hotel might fill up quickly. What travel brands really want to do is actually push people to hotels where there's low occupancy. What we would do at Revlifter in this instance is use our technology, for example, to actually push people to Hotel B from Hotel A where there isn't enough people staying, for instance, in the same location, by pushing things like room upgrades, or maybe free bottle of champagne, or a free meal, or maybe some money off, for example. And it's actually difficult to manage those occupancy rates, unless you're actually managing your yield and your using the data. Someone like Misguided for example, who are masters of persuasive nudging from a social proofing point of view, now social proofing has been around for what seven or eight years I'd say, but there's still a lot of brands that aren't doing this as a very minimum. And this is a really good way to create urgency for a lot of your shoppers that are onsite. Now, these may seem like obvious things, but like I said, these are very much the benchmark of what we would expect as a consumer these days. The challenge that we are really facing is that consumers are very savvy and they really shop around for the best deal. Often I get questions for advertisers saying " if you are trying to drive consumers back to our site for the best deal, they've learned to go to voucher code sites" well, actually that's because they've learned to do that and it's up to us to make sure that we challenge those consumers and challenge their beliefs and where they can get the best deals from. Ultimately as the advertiser, it's our job to market, to them. From an advertiser point of view and the way that advertisers actually do this, if you go to Treatwell, which is a brand that actually houses lots of and luxury sort of bar treatments, basically. So what they do is they actually try and drive people to their app for the best deal. The way they do that is they encourage app downloads on the site or via their partners, which is something that a lot of brands do try and do, but not particularly well, in my opinion. They do this cause they only offer their best deals via the app. They've trained their consumer to understand that they're going to get deals via the app that way. Hotels.com, Expedia, a lot of travel brands do very similar things. Very for example, one of the first brands to really utilise weather actually in house. Now they do it at a very medium capacity, but they're probably one of the better brands actually out there managing this. Being an online shop, they can drive live weather information based upon if it's raining, they'll push wet weather clothing, if it's sunny, then obviously they'll do the reverse. I think this is really important because we always worry about first-party data and rightly so we should, but actually we need to react to the environment around us and sometimes it's best to override deals based upon what's happening with the audience in their vicinity. Lastly Expedia, so presenting dynamic deals on the margin. One thing that we really worry about as brands is actually, how do we balance margins? So again, within the travel space, flights have a very low yield, they have a low margin. Hotels have a higher yield. OTAs are always trying to push packages for example, and similarly for retail brands, trying to make sure you match fast and slow moving products together to try and create urgency for the consumer. Here at Revlifter, we actually see things rather differently. All these things just set a benchmark. But how smart is the data that we're using? Things like social proofing are just showing how many people are looking at a deal. When we're talking to our customers, what we really care about is actually increasing the average order value and increasing conversion, but not just using a median average order value. What we really care about is understanding the tiers of customers that are coming to our site. For instance, with Coast, we use our stretch to save technology to actually tier customers who are spending above the 60 pounds threshold. It's about marketing to those consumers to stretch the basket in the correct manner to them. You're not gonna be able to get somebody to jump from 60 pounds or $60 to 200, it's just not gonna happen. So we need to make sure that we actually make sure that we market people with really personal deals with the products, not only in the basket in real time, but also it has a good yield for the advertiser and pushes them over to the most logical threshold. When we use discount codes, what we often see in the affiliate channel is actually a decrease in average order value. What we're saying at Revlifter is that actually to use these deals in the correct way, and in a really personal manner, we can increase the average order value. We wanted to talk to the customers in a really contextual manner. What do we mean by that? Are they new? Are they returning? It's really important that we make sure that we hit the consumer with the right messaging at the right time. And not just that, but also executing on the right deal. It's really prevalent that consumers that you manage to do that in the right moment, don't necessarily leave the site. They stay on and they will actually click and buy from the advertiser. And I think it's understanding that we worry in affiliate marketing that consumers always go to cashback or voucher sites for the best deal. It's not necessarily the case. But also just from an increased revenue point of view as well, understanding smart upsells. The way that we worked with HP, we actually pushed on the basket deals in real time. So we would increase the conversion rate average order value by pushing the relevant deals to consumers. So to give you an example, if somebody had a small ink cartridge in their basket, we would message the consumer to say, did you know there's a bigger ink cartridge here has thousand more prints. Would you like to replace the item in your basket and give them a very small discount? We got a 90% engagement rate because what a technology can do is pull and push things out of the basket. You'll find a lot of technology partners who are starting to come out, creep out the woodwork now. We're able to manage and manipulate what's happening on the basket to make it a fluid experience for the consumer. This really allowed us to actually treble the conversion rate for HP, and it has done for many other brands as well. I just want to talk a bit about success criteria before we move on. But I think it's really important that we understand exactly what your goals are as an advertiser. I've talked about that at the beginning, a lot of affiliates that we talked to obviously tried to get as close as they can to the brand, but without understanding what's in the basket data layer and capturing first party data, it's really quite difficult to do that, but we shouldn't get stuck on first party data only. We need to understand third-party data as well. Things like AccuWeather, who we partner with, API and live weather into any deals that you are serving just to get stuff on that point really quickly. We actually won an award last year with Vision direct into the Australian market and what we were doing was actually serving consumers in real time with the most relevant sunglasses and eye protection. We're hitting people in high UV areas, for example, and actually telling them they have the wrong sunglasses in their basket. When you're able to help a customer in that way, it doesn't just become a deal experience. It becomes a more personalised experience in the sense that you're helping them find what they really are looking for. From that, you could start to then ascertain insights from your consumer behaviour. I talked a bit about how we execute on deals initially using deals engines. Most technology publishers will use static and dynamic deals, which means they're building out very step scenarios, but starting to turn on machine learning, actually understand specific customer behaviour is really important. What we're trying to do is create an environment for customers where, as I said at the beginning of my presentation, they feel as though there's a personalised experience with your brand and actually reengagement because like I said before we do live in this post GDPR world, people do expect personalised deals and to manage that we need to really understand what's in their basket, what other customers are buying and start thinking about the consumers, not just as a one time thing. How do we reengage them from an existing point of view? That actually starts to be thinking about how to improve incrementality. With incrementality, we think about control groups and test groups. You're probably familiar with working with partners who will say to you we'll have Group A where they don't see any of the technology and Group B who do see personalised technology, let's say onsite, or whether it be through a personalised voucher code page. What we really need to understand again, is what the advertiser objectives are. When we start to observe those current behaviours, we really start to understand how we can start testing and learn it. Testing and learning doesn't really require A, B, C, D type tests. We should be able to test on a one-to-one basis because we're taking things from the basket data layer or data layers that are on the basket page. At least we can then start to actually market to people from the one-to-one basis, because ultimately you can't start grouping people demographically like we traditionally have, because people don't necessarily behave like a herd. We believe that we've moved on from the incrementality idea of just creating control groups. Of course we do have control, repay and control Group B because that gives us a really base level of information. Is the technology engaging with the consumer? We talk a bit about freezing the basket before and post interaction. What did the basket physically look like before consumer actually saw a piece of personalised, technology? And what does it look like afterwards? Are we seeing a different pattern across the consumer base? Working with partners like Commission Factory to understand what the impact across the entire channel is. If we're lucky enough to be able to see cross channel, so all your paid channels, then actually what's the impact across all of those channels? Also only showing incremental offers. I go back to what I've said a thousand times, this presentation, but it's really important that we only show incrementality when it suits you, the advertiser, when it's good for you, good for your margin, and also rewarding customer behaviour. Ultimately, only giving a deal to a customer when they behave in a certain way to release a deal. We work with brands like Puma and we'll be launching Nike soon as an example. With those brands, we're talking about showing consumers progress as to how close they are to getting a deal. So you are almost educating the consumer to help them understand how they're gonna release a code. Searching the web isn't always the best way to find a generic code. Finding our home at the advertiser site is actually the best place for that. So just before I get into personalised deals, I just want to just talk a little bit about how we help brands do that here at Revlifter. I didn't want to get into the sales pitch in the beginning, but I think it's important to understand how we execute here in the business. So what we basically do is build native personalised deals pages, and we drive consumers back on who have say left the basket. What you'll find is a lot of technology partners are now starting to help brands similar to the way that we work, but in additional ways onsite. I want to talk a bit about how I think that is going to help impact the future of personalised deals moving forward. What we're seeing now is a move towards personalised rewards and actually managing rewarded gift carding in a large environment for consumers. So dynamically changing the margins on the rewards that consumers actually receive. Managing that on the basket would be for instance, we have a consumer who has a pair of trainers and actually educating them that if they add trainer protection and white socks, you're gonna potentially reward them with either a deal or even a dynamic gift card. We all know that Amazon's risen to the ranks over the number of years, and consumers have learned to go there to find pretty much any product that they want. What we're finding is a shift in retail behaviour that they're trying to wean themselves off Amazon. I guess equivalence in most markets would be creating or finding at least deals pages on the advertiser sites. However, most of these pages are static. So where we're try and help consumers at Revlifter is by creating these dynamic deals that we are driving people back to once they've left to go to Google to come back on an SEM PPC basis. We're also seeing a rise in voice searches as well. I saw a stat somewhere there by 2022 50% searches will be via voice. When we start thinking about that, that's massive. Are we moving with the trend? You'll expect black Friday and also Christmas periods this year, there'll be a lot of people using their Echos or their Google to search for deals. Also brand direct to customer, the conversations I'm having on a daily basis with brands is how do we get customers direct to site? Brands I'm thinking of are Nike, Puma, even HP, for example, a lot of electronic companies are trying to move away from relying on resellers. I remember 10 years ago taking co-op funded money and not being able to put the best deal out, even though Intel had paid for a HP campaign, because a reseller had a sale. And now we're moving beyond that world where we're worried about those relationships. Also use of geolocation, we're pushing people into store. You look across the globe, loads of high street retailers are closing their stores down, moving to online. Some are even going bust or into administration. Actually driving footfall in store, but also driving high street traffic to actually shop on their mobile, offline to online as well. How do we manage that? How do we manage the flow of traffic? Making sure that we actually capture data that wasn't digitised before. A lot of brands now, their technical houses or their technical functionality that they have for storing data is vast. How do you manage that? We're able to skim on the top of the data and actually understand how we manage that, because we have partners that can process that data. If you look at partners like Adobe, for example or the big media houses, and actually here at Revlifter we're able to pull a lot of not only the basket page information, but also electronic points of sale as well from the advertiser, that gives us a new dynamic on the way that we shop. If you look at brands, like for example, The Iconic within Sydney, for example, but delivering goods within the hour and then be able to send it back, so you can actually get ready for a party in the evening. There are very few brands that are able to facilitate that, but that's a huge leap forward. Open banking also, so thinking about what is it you are buying with your Visa actually using that information. I remember eight or nine years ago, having a meeting with partners like Visa who were trying to understand how we use that information. We're in a wealth now where we can access that and actually personalise deals for the consumer in real time. At the beginning of my presentation, I talked about consumers who will take back control. I still hate people saying people are worried about the security, they're not willing to give up their data, we need to be really worried. Yes, of course they are. I certainly am, we all worry about that. They also understand in this post GDPR world that actually their data's a commodity. We should understand that they're expecting personalised deals. Single points of application, so that's basically, when we talk about, when we have to refresh a deal, we usually have to refresh the whole basket or the page. Here at Revlifter, for example, we can refresh offers on the fly on the page itself. When we drive people back to the native voucher code page that we built we're able to actually refresh those offers based on what they've had in their basket previously. Similarly, on the basket, that's also true. An FMCG will catch up with retail, Unilever, for example, spend more than Tesco here in the UK, and you'll see that with brands across the globe as well. There are brands that have untapped budget, which they're willing to move into affiliate marketing. Traditionally, affiliate marking has always been one of those acquisition channels or people think it's just driven by deals. If you think about it, affiliate marketing really is a payment metric more than a channel. We house as an industry, all of the other paid methods of driving traffic, we just do it on a performance metric. Certainly what we're seeing also is a rise of ethical rewards, consumers are very savvy. If you just look at the rise within the younger generation with the fear towards things like climate change, for example, ethical rewards is a natural fit for those types of consumers. Brands like Toms, for example, who are gifting shoes to poor parts of the world. I think we'll see a real shift. In summary, just a few points that I wanted to make really, before we open for questions. One, I think the brands that take on board these personalised techniques and actually understand that their consumers are expecting personalised deals like the ASOS, The Iconics, the Boohoo they're the ones that are going to succeed. The Iconic, for example within Australia, they're the brands that are gonna succeed and they're the brands that are gonna fight against the downfall of the high street, which we're seeing, because at the moment, the high street isn't adapting. Number two, the high street does not adapt and understand that it needs to start driving or using data to attract its customers and create different shopping experience, then they're going to really struggle. I really believe that brands that have their own goods, for example, are really gonna start to push to drive customers direct, and that we need to be very wary that brands like Nike and Adidas for example, are gonna be trying to drive more footfall, not only in store, but also online. If you look at things like Nike Labs, for example, through how they personalise goods for their customers in a really unique way, are starting to see that massive shift. I just want to leave you with one final thought from Jeff Bezos. if you think about it, this is over 20 years old and actually it's really prevalent today. I have 45 million customers. I don't just want one store. I want 45 million. If we're gonna think about how we're gonna fight against the likes of Amazon who are personalising and driving forward the immediacy that customers are expecting through delivery within the hour, we need to start taking on board that customers expect personalised deals in real time. I just want to thank everyone for listening to me today, I'm just gonna hand back to Amanda.