Performance Series Episode 3:
How to Win Back Customers (The Right Way)

Thanks for joining us today for the Amarsys webinar series. I'm Nick Odom, director of technical adoption, joined today by Colin Clark, one of our enterprise sales reps. Colin, thanks for joining us today. Yeah. Thanks so much for having me on the show. Yeah. So we, we brought you in today just to kind of talk through, we're wrapping up our our life cycle marketing series for for q one, and we really wanna talk through, like, winning back, defecting, and inactive customers. And I know that's one that that we've kinda talked through over the years. You know, I know you've worn a lot of hats around here. So just some background around retention marketing and things that you've seen over the years, before we jump in. Yeah. So, I've been with the Marshalls for about five years now. I started as a strategic consultant. So I was working with a lot of existing Marshalls clients. So actually, you know, going into the platform and building a lot of these types of programs. I've built dozens of different win back type programs to to help customers use their data and use all of the, content and channels that we have available to, to impact customer lifetime value. I did that for the first couple years that I was at Amarshez. Also worked as a solution consultant for a couple years. And then for the last two years, I've been, an account executive working with, our larger enterprise deals. So new business sales, working with big, retail and ecommerce businesses. You've really kinda seen the the full spread of of different use cases, things that different brands are trying to do, and, that that's awesome. I I know we've, we've worked on a lot of stuff over the years, and I've always enjoyed talking shop with you. So really excited to jump in here. Just before we go, like, what's what are some things that you think about win back? Just like a few quick hits, like, when you're when you're first starting a discovery process of if we should run a win back or not. Like, what are things you're looking for? Yeah. I mean, I think there's there's really kind of three big things I like to target. So one is to really understand the defecting customer window. Two is to have a content strategy that's gonna keep things fresh, keep the the messaging relevant. Mhmm. And third is to think about some different ways that we can make the campaigns more complex. So whether that's adding more channels, adding additional segmentation, adding additional AB testing, we're really optimizing these campaigns, in the best way possible so that we can get the most out of them. No. I love that. I mean, I I think that really goes back to what we focus on the existing side. It's like, these programs aren't, like, set it and forget it. You know, they it's nice to get something into place. It is a machine that's just gonna be able to run. But at the end of the day, if we're not going in and optimizing, adding in new channels, you know, building in new dynamic content or new branches or new timings, like, it's gonna get boring and stale for our customers. And that's where we start to see diminishing returns on some of these programs as, you know, okay, this is ran for a year and a half. The same people, you know, the same people have seen this two or three times potentially. Yeah. Absolutely. And I think that that's really important is to understand what does your defecting window look like. So Mhmm. One of the great things about VMars is just as a platform in general, this gives you that ability to import, you know, anywhere from two to maybe ten years of historical data to run all different sorts of machine learning models on that data to understand the full picture of what a defecting customer looks like. So what we'll see is, you know and this is almost universal amongst all brands that we work with, is at some point, you'll start to, analyze from the data that customers are kind of reached a churn phase. So that'll be, you know, essentially, after a customer is an active customer, they become a churning customer. And then, you know, maybe for two hundred or three hundred days after that, there's still a chance to win that customer back. So we call that the defecting customer phase. Yeah. And, and what's great about that is it gives you the ability to not just create, you know, a a static win back program, but it could be a series of different programs based on the customer's engagement, based on the number of days since last purchased, based on purchase categories and purchased amounts. So, you know, really understanding the customer and and customizing that journey so that you get the most, ROI out of those types of programs on the, you know, far end of the spectrum. Yeah. And I I look at that a lot with with customers that are looking at these types of programs as we we go into our RFM advisor and actually see if the timing of the the segments are really right. You know, especially coming out of twenty twenty where so many more people were buying online and purchase frequency in general for a lot of brands shrank. You know, they were maybe running or looking at running a win back at that, you know, let's just, for example, say say a hundred and eighty, two hundred day mark. But over the course of the last couple of years, they've actually drifted down into that, like, one twenty, one fifty mark. And, you know, that that timing is so critical for for the program and the content because if you're, you know, two, three months off, they might have gone to purchase some more. I mean, in You know what I mean? That's an important call out is that if you're doing the right things, your purchase frequency will compress. So your active customer window should compress a little bit. Your defecting window should cut compress a little bit. Mhmm. And, and adjusting that timing, understanding, and and evaluating what the churning buyer looks like. Right. Very, very important for for optimizing churning. Like, trying to get some of the the brands that have maybe historically looked at these graphs and known it for, like, being the hundred and eighty day mark, you know, trying to get them to shift and say someone's defecting at a hundred and twenty. When if they're running some of their, you know, like, even finance reporting off of some of that, you know, sometimes they'll freak out a bit because you're you're making that window you're you're making that group potentially bigger for a time period. But at the same time, the content is timed a little bit better, and that active group is really, like, more of a hyperactive customer segment. And we can start to align, you know, new products, new content, like seasonal content to the beginning of that defecting cycle. And to your point, you can have multiple win backs across, you know, that entire thing. You have to go at the beginning of the cycle, the middle, the end. Mhmm. Absolutely. You know, as far as, like, personalization, obviously, you mentioned, you know, getting some recommendations, the timing, this kind of stuff. What other things do you look for when you're building out some of that dynamic content? Yeah. I mean, if if we think about, see, some of the other webinars that that we've done, talking about welcome series, post purchase, we're always trying to get more data on the customer. So whether that is web browse data or email engagement data, past purchase product data, all of that can be used to feed into a content strategy that allows you to you kind of send the same email over and over, but it renders differently every time. So all different recommendations, or we're doing banner swaps in a dynamic way that are based on, you know, data that I've given the brand or data that, the brand has gathered just for my browsing and purchase history. Mhmm. So, so there there's lots of things we can do to kind of scalably make content render and give the experience that things are different without, without having to They'll lay out the frameworks about the same, but, like, you know, you're you're putting in a new banner. You're you're building out some new grids for the racks. You're maybe changing if you are doing run an incentive strategy. You're you're changing up the incentive based on lifetime value. But, like, it's it's almost just taking things just a little step further because without doing those, you're you're just gonna end up sending the same email over and over, and it gets a little boring for your your constituents. Absolutely. Yep. You know, you'd you'd mentioned, you know, maybe even bringing in some other channels. You know, I know we've talked a lot around program complexity, like, testing into some of these things. There's a lot of, you know, even new channels that people are activated on nowadays. You know, how how does some of that play in when you look at social media, SMS, push, like, all these different things that we're playing? Like, how can you maybe effectively position that in a in a program like this? Yeah. I mean, so it starts with, with having a good permissioning strategy when you're onboarding a customer. So understanding that once a customer signs up for email, whether to get an offer or whatever, maybe we need to permission them on SMS, permission them on browser push. Like, long before this went back across. Yeah. Yeah. So that like, that those foundational parts of the strategy. And then another thing would be would be mobile. I mean, mobile is is huge, with, push notifications, in app messaging, all that kind of stuff that you can coordinate with a campaign as well. But but thinking about the fact that any touch point that I have with a customer that's in the digital realm can really be personalized, especially with the Marsys, you know, through, a very easy mechanism of just dragging and dropping personalization presets, dragging and dropping recommendations so that every single piece is in the channel that the customer expects. It's content that's relevant to that customer, and it's tied back to a business KPI. So in this case, for WinBet, we're typically talking about increasing customer lifetime value. And, really, in a roundabout way, it's kind of, decreasing customer acquisition as well. Because if you could sell more stuff to your existing database of customers, then you don't have to spend as much money money on acquisition on the front end because you've got all that revenue coming from existing business. So it's looking at those two pieces and those KPIs and then having all of your strategies kind of, act like a flywheel to drive up, up and down the funnel. Because this can really be a a huge mechanism for reactivation. I love the the call out on the the new the new customer, the cost of acquiring a new customer because we'll see that on, like, really two ends. We've got, like, those emerging, like, newer brands that have focused so much on acquiring new customers. They're out, you know, on social media all the time. They're acquiring, you know, building their list. You know, you've also got this other kind of interesting subset of of customers that we work with that are maybe a legacy, like an older, more mature brand, but they're just getting into direct to consumer. You know, especially coming off of twenty twenty, so many people that historically had sold through third party sellers, you you know, started selling their own products online and building that that rapport. Both were focused a lot on acquiring new customers. But as we've gotten into twenty twenty two, we're we're now potentially at a new buy cycle for some of those and, like, focusing on a really effective retention and win back strategy could really save them a lot of dollars on upfront cost. Yeah. Absolutely. Absolutely. You got that data. If you're collecting that data, it's really just a question of how do we leverage that data to actually message customers effectively and make it have a business result within the business. You know, and that's that's a great that's kind of a great segment of my next question. It was just around, like, testing. Like, you know, you're you're building out these things. You're putting effort there. And, like, sometimes, it's a hard sell to get, a a client to go from, you know, historically doing batch and blast and just doing kinda list distribution stuff to to focusing on some of this retention cycle and, like, building out some more robust, you know, flexible content models. How do you test into it and say this is more effective than that? Yeah. I think there's there's two different pieces to testing. One is your traditional AB testing. So testing pieces of content against against each other. Whether it's subject line or banners or whatever it may be. Yeah. Which is super, super important. But the other piece that, you know, I think is equally important that you could do within tools like a Mars, like, this kind of next generation automation tools, is to also break out customers by your different attributes. So we we start by having really, really good customer intelligence. So we understand, where any problem spots may be. So we so we for example, we might see that we have lots of customers that buy once and they never purchase again. Well, that might be something that we wanna branch off and test different options just on those one time buyers that aren't purchasing. We may have some top categories that are, are very, very prone to churn. So let's take those top categories, break them out, and do some specific testing around those specific cohorts. And then also, monetary is super, super important. I mean, if we go into any dataset of customers, we're gonna see super high value customers that are also churning. Those are important customers to retain. That's where a lot of your money's gonna be tied up if you lose those customers, to, you know, either, you know, not purchasing anymore from you or to a competitor. So that's super, super important in those cases to have an intentional content and testing strategy to optimize the the win back process for those high value customers. Yeah. And I I think you all of that's, like, super interesting. The the one that I I know me and you worked a bunch with back in our strategy days was that, like, the the top churning categories, but also, like, what categories people are likely to buy next Mhmm. Sort of thing, where, you know, if they buy into this category, you know, you might be able to position either a product similar to that or, you know, an accessory or, like, if they bought maybe, you know, shoes one season, they were likely to buy jackets the next or, you know, whatever that might be. How how much of that kind of, like, analysis is you're, like, structuring out some of this content even if you're within, like, a post purchase series? Series? Because I think that's just kind of long tail end of this. When you're looking at some of those affinity charts, I mean, looking at, like, first purchase to second to third, I mean, how much of that do you really go in and map out and, like, try to really understand the customer and some of these unique customer journeys that kinda come out in the data? Well, I think, first of all, first of all, it's important to think about how you're gonna scale all your programs. So so something that that that level of complexity is not necessarily right for every single brand. There's probably value you can get out of every brand, but if you only got a two person team, it's gonna be tough. If you do have, and and the other thing is, you know, within the Marshes platform, it gives you a lot of those charts and graphs and and analytics around product analytics, customer analytics, life cycle analytics. So, so there's a lot of data available, but that I think, really, when we look at, you know, the the charts and graphs and, customer analytics part of the Amarsys platform, it it informs you of what to do. Because you'll see, oh, we have lots and lots of customers that have affinity towards winter coats in the winter. And and then you can, you know, take those campaigns, make them dynamic so you can, have them be kind of evergreen all year all year long. Yeah. I mean, I think that's a that's a great point of, like, how far is too far for, like, return on value? Because, like, there's there's some brands where I think that's, you know, maybe an easier task to get to. Their catalog's not near as big. They've got very clear kind of purchase lines. That makes a ton of sense. There's also, you know, different product recommendation logics that we can lean on. There's a post purchase logic to, like, you know, pair up to things they may be bought or, you know, things that they're they've been browsing recently. And there's some of the whole let the machine do the work kind of thing in there when you're trying to, you know, define some of this incentive strategy and how to scale it. Awesome things. Just kinda looking at wrapping up here. I mean, what's what kinda, like, checklist items would you have if you were going and talking to a new brand? They've got a few programs in place. You're looking at, you know, aligning this. Like, what are some, like, quick hits that just anyone could kinda go in and, like, you know, look at their information and figure out if when back would be right for them. Yeah. I mean, again, first is gonna be the defecting life cycle. So understanding, at what point does a customer start to churn, and at what point do they become inactive, and what does that look like in the middle so that we can activate across that? You know, if they continue to not engage, having mechanisms in place so that we can bring that customer back. Again, dynamic content, you know, having it be a fresh experience every time I get entered into a win back flow. Yeah. It's not super, super hard to do, especially if you've got recommendations integrated in your platform like we do with the Marsys. And then just thinking about, you know, not having it be a set it and forget it type of program, adding additional, cohorts, different branches, different AB testing, optimizing over time so that we can get a better ROI out of each program that we're running. Makes a ton of sense. Well, I really appreciate you joining, and we'll hopefully have some conversation like this in the the future. For any of you out there, if this is something that you guys are interested in in learning more about or even looking at your own database, reach out to your CSM or or an Amarsus representative. We we'd be happy to kind of dive into it with your brand and see, what types of things we can maybe get in place in in your account, if if that makes sense. So I think we're gonna cut to live QA here, and then, I think that's a wrap. Colin, thanks so much for your time today. Yeah. Thanks, Nick. Yeah. Hello, and welcome to the post webinar q and a. We'll be monitoring the Q and A chat room. So if you have any sort of questions you'd like to send over, feel free to post it in the Q and A chat. Okay, nothing so far. We've also got Josh queued in to the Q and A here. Josh, how did, what was your take on the episode there? Yeah. Hi, Nick. It was really great. I thought, Colin brought a lot of really interesting, you know, perspectives, to it. And and I think a lot of it was that really appealed to me was keeping the content relevant for, like, year round and and and seeing what kind of changes are happening to your, you know, product affinities or is the actual time frame of it affecting or inactive, you know, group, is that changing? Basically, it's just, you know, thing of of once you set it, it's not it's not, you know, gonna be the same forever that you it's it's constantly needing to be sort of evaluated and and updated to to adapt to the the customer base and what they're interested in, since that's always gonna change. So Yeah. Definitely constantly evolving our our content strategy. So I've got a couple of questions in now. Any video recording to share? We we do have, we will be releasing a recording of the video. There's also our our previous episodes from earlier in the year available for download as well around, like, the welcome and post purchase series. But in the coming days we'll have a recording for this one as well that will go out to all the registrants. Another question coming in here. Thanks again for the awesome session. Thank you. Thanks for joining. And then do you have a trial platform where we can check out the features? If you're not necessarily a Marsys client, we can we can definitely connect you with someone that can, definitely demo the platform. And then if if needed, we can we can potentially figure out, some sort of sandbox that we can we can look at. But just reach out to your representative, and we can we can check it out. If you are an existing client, you know, again, kind of the same thing. Reach out to your rep. We can, if it's new features that we're we're interested in, we can definitely make sure that you get the the information you need regarding that. Yes, Josh. I mean, definitely got to just keep it fresh, right? Like, that's kind of been the theme of this whole quarter is just these kind of, what I think historically we've considered evergreen campaigns, like we just we've got to keep evolving them, we've got to keep adapting to the needs of the market, you know, keep building out more specific content for the products that we're selling and just, just keep it fresh, right? I think that's kind of the thing is like don't let the content get stale. We start to see diminishing returns in some of this stuff when it's sat for a while and people have seen it multiple times. So I loved Colin's perspective on just testing into that kind of stuff and just ways to evolve. Yes. Can I go ahead, Josh? No, let's hit the question first. Yes. So next to offering a discount or giving extra information about products, what other tactics or incentives can be useful in order to regain customers' attention? You know, it it really depends on on the company, but there there's so much, kinda at stake. I mean, I I'm not necessarily the first to lean into an incentive, for this kind of thing. I I do like showing new product new products and, you know, trying to put, you know, that sort of stuff. But I also see a lot of traction with companies that maybe put lifestyle, imagery and and lifestyle kind of messaging into some of these programs to really, you know, show how the product is being, like, used and, like, you know, kind of building some interest in that, you know, say you're an outdoor brand, you know, and you're you're, you know, coming into a new season and you're just kind of encouraging or teasing people to get out and do the activities that they would need your products for. So it can really depend on what it is. But Josh, what other things would you maybe think of from a content perspective outside of discounting, or just additional information about products? Yes. There's a lot of different kinds of incentives that aren't necessarily just a discount. If you have any sort of, like, loyalty type of of program, that that you feel like would would help win these customers back or, or if there's something that's kind of setting your, your brand apart that, that would entice someone to to go with you over over somebody else? And, you know, do you have something, like, that's that's evergreen with your product discounts that maybe isn't a discount, but maybe it's, like, free shipping? Or, you know, you can sign up for, you know, certain, you know, like replenishment emails if you're, wanting to, make sure that you're kind of completely up to date with with products if if it's if it's that type of business. So there's there's a lot that you can do that isn't, you know, it doesn't have to be a certain percentage off or a certain dollar amount off, but still gets into the the the customer's mind. I love the call out on like maybe some seasonal content that makes sense for potentially returning customers. Kind of leads into this next question. Do you advise on a win back series? If yes, how long do you keep trying to convince your customer? Would you recommend data cleaning if the customer doesn't respond? You know, and and I think there's really two ways to look at this. You know, when we're talking through these win back series, it it's really around, the customer's buying cycle, you know, and and sometimes it gets mixed up with, you know, like an engagement, cycle. So in the case of, a win back, and this is also where we like to split, you know, defecting from an inactive because it's really starting to set a stage for, like, how far defective they are. And so if you're looking at, like, what kind of content play we have for a defective audience, you know, this is really the the the time where we're thinking, you know, they just started to show these signs of of churning. And and that's where, you know, plugging in new products or lifestyle imagery, you know, like, we we can hold the incentive back, you know, at this point. Like, we're just trying to to generate some interest with a customer that we previously had. You know, as and they start to drift into what, you know, in a Marxist world, we would consider an inactive customer. You know, that's when we can start to to be a little bit more, you know, maybe concerned that they're they're not going to return. And and there maybe is a case where it makes sense to maybe sunset, a customer, You know? But I also in that case of of cleaning the data, I I like to keep engagement in mind because we've got an active, you know, email subscriber that continues to engage in email, continues to to look at the website, it just hasn't made a purchase. It's maybe someone we wanna clean off our list but maybe not, you know. So there there's just a couple of, different different spots that I would consider, you know, looking at before determining that we should just totally wipe someone off of our list. Yeah. And you can always, you know, have like a, at the end of your win back program of of someone who hasn't engaged, with that program or or maybe has engaged but, like you said, hasn't made a purchase. You can easily put all of those people into a contact list. And then if you wanna suppress them from certain SIMs, then you have that, option. But then, you know, say Black Friday rolls around or you have, like, a really interesting product that is, you think can kind of have mass appeal, then you can bring that list back in, and that can kind of be another way of trying to, win them back. And you haven't, you know, completely removed them from their database, but you've maybe just secluded them from some of the more day to day something? I do I do like that that call out on suppression because I I think that's something that people miss a lot and especially when we start to see, you know, lists getting pretty large and, you know, maybe even encountering some deliverability challenges, you know, in that in that batch, you know, we all we all do it in the batch sending, you know, cadence that we end up in week to week. Like, it it probably is important to to maybe suppress some of these people that haven't engaged. Just give them a time out period. You know, it's not necessarily helping us with someone that's continuing to not open our mail or not click our mail week to week. Yeah. And then maybe Unless you around go ahead. I was just gonna say, yeah. And then maybe you, you know, add on a new channel that you didn't have before and you thought maybe that would, be something of interest to these, inactive customers and then you again you always kind of have them in your back pocket instead of having just kind of deleted them completely. Which is a perfect segue into this next one. Would you recommend using a segment or an audience automation center audience, for Sierra Sierra Mads? So for those that are maybe not familiar with them, our Sierra Mads are our connection to to the Facebook and Google networks. And this is really honestly, it's an interesting it's an interesting question because I I think it you know, I hate to say it it depends. You know, a segment is maybe depending on what the parameters of that segment are. I I could live in that segment, you know, each day for a year, and and I I might continue to be targeted on social media because of that. You know, but you could also put parameters on there, you know, depending on what's what exactly that segment is. But just knowing that that segment will refresh every day and and overwrite whatever list we've got, you know, ported out to whichever network we have. You know, so that's something to keep in mind. The automation center, on the other hand, is interesting because we can force the time frame in which they live in that audience. So we can maybe bring them in and in some ways it's still based on a segment, you know, say we've got the win back and we refresh our defecting audience every day, we send them an email, we see if they respond. If they didn't respond after a couple of days, we're gonna drop, you know, maybe our high value customers into, you know, the the the, DRM ad audience in in the in automation. You know, from there, we just want them to live in there for a week, and then we want to start to suppress. So in that case, we've got a little bit almost more control based on the timing of them entering the program. So again, it just depends on the needs of the audience. It might be something where we just always want our high value defecting customers to be decent over Facebook and, you know, having that outside of the automation might actually be less complicated, you know, but it depends on the needs and, you know, it also depends on how large the groups are. You know, we want to as much as we wanna use our first party data to to influence and and honestly optimize our spend on on these social networks, we also have to make sure that our groups are are large enough to support the social network. I think it's like two thousand two thousand is the minimum of group size. So sometimes with the automation, you know, if if that automation is not gonna support the traffic to push enough people into that group, you know, that's another thing to consider. And that might be, you know, a spot where we actually look at more or less automating outside of the the actual workflow, and building a segment that's kinda combined with some other other audiences to make sure that our our group is you know, significant enough to to be targeted. Yeah. I think the the group side is a a really good call out because the yeah. Like, the automation, you can have so much control over and you could do, you know, a, an a b split with the the the audience and kind of go email with one size and, and CRM with the other and see how that compares. But you yeah. You would definitely need the, like, a decent group size to start with if you're gonna be kind of splitting off into different paths. Alright. So the next one, guys, you know, how how can we determine the right moment to push a message after a customer is becoming an active? I've been taking a lead on these. You want you want this one, Josh? Yeah. I mean, I think that this is what the, the the customer life cycle is is, is really good for. And I know that the the immersive platform has a advisory tool that, says, hey. This is what we initially had as a customer becoming defective, and a customer entering that inactive phase. And then but based on the purchase data that we're looking at and time between purchases, here is kinda like the most up to date recommendation of of of what the platform is seeing as as that start of a of a defecting phase. And I would always, say like, I think was was mentioned in the the, chat that you and Colin had was, try to always push that message to get someone to come back as they are entering defective and they're not that fully inactive yet. So that it's not, you know, it's not too late and that they're not fully disengaged by the time that you are are sending that message. Yeah. I think that's a great call out on, like, the advisor and just kinda optimizing the groups. You know, I I think there's there's some other points in there too around, like, you know, maybe using some time optimization. You know, if I'm looking at, like, some of my triggered programs, you know, my abandoned cart browse, I kinda want those to go out at, you know, relevant to the the event that occurred to trigger it. But with these kind of look back programs that are, like, you know, recurring segments, you know, the timing of the day is maybe less crucial because, you know, they're they're kind of, you know, going going stale. So we're we're, you know, pushing this message proactively to them. But then depending on the time of day we send to them, you know, leaning on something like send time optimization, can really, you know, beef up some of our open and click, you know, mainly click rates, to to make sure that we're we're kinda hitting the inbox at the right time. But the other piece, you know, around, like, the the audience segmentate segmentation is, like, you know, our our AI product, which is, you know, basically these same segments, but on a one to one basis. So, like if we're looking at a defecting audience, the AI segments are going to look at that audience in a one to one fashion. So, if Josh you're considered defecting after like twenty five days, maybe on defecting after sixty days, you know, this program is going to be tuned up so that you get pulled in earlier than me, you know, and the timing is good. And I think what's been crazy for me as I've worked with that that product is is taking the exact same content that we used for a standard defecting win back which is kind of based on the group taking that exact same content and plug it in into these segments and then all of a sudden we've got click rates going up. And the only thing I can attribute to that is the timing, you know, because, you know, the machine kind of determining, like, when when these people should be pulled in. In the case of, like, me defecting at sixty days, you defecting at thirty, the whole audience looks like on average we're all defecting at, you know, fifty days or thirty forty five days or something. We've missed you, right? Like you've maybe gone and bought that product somewhere else because you actually needed it two weeks ago. But just tuning that up and looking at it on a one to one basis can really, really optimize the timing of when those messages drop. All right. Couple more here. How do you make sure the win back campaign doesn't lead to one single purchase that actually engages them long term? Now I think that really comes in when they do make that first purchase, right, like we're this is trying to drive to a second purchase. Everyone that falls into a win back has had at least a single purchase. And I think there's a couple of things to look at in there. Like do we have some leading indicators on categories that do drive high lifetime value? Are we kind of filtering out our past purchases screen if you are an Amarsys customer? And kind of filtering out and driving that lifetime value filter up and up and up and seeing how many of those categories and products kind of exist and like maybe there's a leading indicator on a potentially high value customer just based on what they've purchased first. But really and I think that's where the win back can come in play to try to keep that moving, but that's also why I don't want to lean on incentive strategy so heavy because that turns the brand into just one coupon after another, but instead leaning on maybe new products, product activation, lifestyle imagery, like just what that product kind of means for the customer, why they might need it, like really selling the product itself and not necessarily the discount, that can really build a lot of brand equity and like start to build that loyal fan base. But that's it really depends on like what the product is, why someone is coming to your side over another, if we've got competing products in the market that they could also be buying from. And that's something that our team will look at a bunch with like doing competitive analysis and seeing like what kind of content other people are sending and where we can differentiate ourselves to make sure that we are putting brand equity in the programs and the content we're sending now. What about you, Josh? How do you kind of think about that first to second purchase and like making sure that it's actually for the long term and not just a quick win? Yeah. I think there's kinda like what you're saying there's, I think two big pillars. One that's just these are the type of products that, are that we've seen are successful at turning someone from a first time buyer to an active buyer. And it's just like the the the way that those, products are that is is kind of has mass appeal to everybody. And then there's the recommendation specific to, well, what was the first product? Is there something related to that first product that's, an easy recommend? Is it, you know, an accessory to that product? Is it, you know, something like that? And then in terms of of, you know, win back, if they've become defecting, if they've become inactive, then, yeah, you don't wanna just throw a a a huge discount on there to get them to do one purchase and then they become inactive again and they you know, six months goes by when they get another, discount email and make another purchase. You know, that doesn't actually keep them engaged. So yeah. It's like, do you have a, you know, a newsletter? Do you have, anything, you know, that you're actually, like, educating the customer and, actually, like, forming a relationship with them where they're coming to you for products hopefully, but also maybe information and, and maybe relying on your brand for those recommendations that, you know, maybe they don't, know exactly, what they want yet, but they know that you kind of have a history of of giving them, solid recommendations. And so then they are more likely to go to a site that they have history with, versus something that's new that's just gonna give them, the most popular product. Building that trust is is super critical, you know, in the heavily advanced e com world that we're in now. I mean, it's, you know, to your point, like, people want to go back and shop the same places that they did. There's some loyalty in that for sure. Kind of last question here before we wrap up, it says, is there anything you can share around the deduplication process of the platform? Is it very detailed? How relevant is it within the win back process? There's a couple of ways we can look at that. One, we do have a unified customer profile that sits in the platform and that's what ultimately houses our customer information. And so the ways we import into that, you know, kind of depend on how strong our DG processes. Is that something that you would go through during onboarding just to make sure that our keys are good and, you know, our we're we're not importing and creating duplicates, you know, with our contact sync? And then as we look at the win back process, you know, we we bring in, you know, your sales, you know, during onboarding, we bring in, you know, two two to five years of of sales history, and then every day, we'll bring in a delta file after that. And so that's that's like a one to many relationship to your to your contact database. So we're merging, you know, with a consistent ID, your your sales back to that customer profile and then determining, you know, what how many purchases that person's made. We're calculating you know, lifetime value. We're calculating their their category and product affinities. And it's also ultimately what powers, you know, our programs like our post purchase and our win backs and, you know, a number of other, programs that we have built around the life cycle. It also enables you to do some advanced, you know, filtering for, your, you know, your just your badge sending, honestly. If you if you were looking at, you know, what what affinities or past purchases someone's had, you know, that you can more effectively segment, your list potentially if you've got, you know, an email going out around a particular product. So, we we definitely have duplication, you know, kind of built in to the platform and we try to make sure that during onboarding we're, we're building that that contact sync as robust as possible to make sure that we're not creating duplicates and, you know, people aren't necessarily falling into, you know, this this program errantly when when they've already had another another purchase since then. Well, I really appreciate everyone's time and questions. Hopefully this was impactful for you and your business. Thanks again for joining us and we'll talk again soon.