The Death of Retention Teams

The retention team is like a traditional 4 to 6 person retention pod. I think it's dying.

Brands I'm working with, I guess, that used to have retention marketers, multiple email other folks, designers, copywriters, maybe even developers or QA folks, they're starting to shrink the team.

It's an 8-figure D2C brand in the beauty space, and they had a 6-person retention team, like a manager, 2 coordinators, a designer, copywriter, and even an SMS specialist. Their team is now just 2 people. And it's just a retention manager and a coordinator.

The new bar is AI plus a smart operator versus a tired team running an ungodly number of campaigns a month under immense stress and pressure.

The work isn't going away. It's just consolidating into fewer hands. But getting really good at directing AI, getting really good at strategy, getting really good at protecting your brand voice, that's the job for 2026 into the future, and you're going to get paid for it.

From the open to the click, they're the masters of the flow. Just watch your list grow and your leads turn to go. Did we retain you so far? Or do we have to extend it? If you want results, you know you gotta send it. Yeah, you know you gotta send it. Yeah, you know you gotta send it. Yeah, you know you gotta send it.

What is up everyone? Welcome back to Send It, where we discuss anything and everything retention marketing, specifically in email and SMS marketing and e-commerce. My name is Jimmy Kim and I'm here with my co-host Chase Diamond. All right, Chase, I want to start with a hot take here, and the hot take I have for today's pod and what we want to talk about is that the retention team is like a traditional 4 to 6 person retention pod. I think it's dying. What do you think?

Dead, actually. You think so?

I don't know, man. I don't want to get that edgy because I say that because I've been watching brands. They've been restructuring for the last 12 months and the writing seems like it's on the wall.

Yeah, I would agree with that. I definitely feel like there's a lot of restructuring happening. You know, the brands I'm working with, I guess that used to have retention marketers, multiple email other folks, designers, copywriters, maybe even developers or QA folks, they're starting to shrink the team. So I guess you're right, a lot more AI is starting to come into play. Is that kind of where you're heading with this?

Yeah, I think that, you know, it's interesting because it's not like people are sending fewer emails. They're actually sending more ultimately, and they're making more revenue per email these days. And that's the part that's a bit wild, but let me give you an example, man. So I had a brand that I was working with reach out to me about 3 months ago. It's an 8-figure D2C brand in the beauty space. And they had a 6-person retention team, like a manager, 2 coordinators, a designer, copywriter, and even an SMS specialist.

Nice. So on paper, that's kind of like the setup that people run and it sounds like they were really strong on paper, right?

Yeah, on paper, yeah. And their email was doing about 28% of revenue, so it was pretty decent, but it was not great for beauty in my opinion, but it was decent enough. And they reached out because they were looking to optimize their program.

Got it. And, you know, I guess I have a guess where this is going, but do you want to tell me?

Yeah, man. So I went in and I audited things and, you know, as we've watched them progress, they're already increasing their revenue. I think they're closer to like 36% now of their email SMS. Up from that 28%. But the thing that I thought was kind of crazy was their team is now just 2 people and it's just a retention manager and a coordinator.

Oh shoot. So they cut back pretty substantially. What about the remaining? Was it other poor people or so?

Yeah. And look, they didn't get fired. They got reassigned to other parts of the marketing team and one even left voluntarily. So it kind of worked out and I think they might have let go of one too. So, you know, I think the brand is doing better than ever and it's, basically done in retention. However, it's just an interesting story that I'm seeing all across the board right now.

Got it. So what you're kind of getting at is this is becoming more of the norm, or at least slowly becoming the norm versus the exception, and that more of this is probably going to happen as the years progress.

Yeah, exactly. So look, today I wanted to do a pod and I want to Not only just break down what's actually happening, why it's happening, and more importantly, basically talk about how to operate as a one-person retention army in 2026. A name, title, a pod somewhere will be that. And so whether you're an in-house marketer or trying to keep your job, or you're an agency operator or founder doing it all by yourself, I think this is going to be a good pod for you guys. And let's be real. Look, this is uncomfortable for some people to hear. That's going to be the truth here, but I'm going to get DMs after this episode probably, and people are going to have their own takes, which is cool. And I'm always open to talking about it. But you want to talk about this thing?

Sure, let's do it. Sounds like it's going to be a little bit spicy.

Yeah, I think a little bit spicy, but it's the reality. And we don't hide things. We don't sugarcoat things here. We want to make sure that we're helping the market out. So let's start with the why. Why are retention teams or why is the retention team, as you know, getting restructured out of existence? Any thoughts here?

Yeah, I guess like, I think it comes down to a couple forces hitting at the same time. And it's not one thing, it's probably a combination of 3 different things that really is why I guess we wanted to talk about this today.

All right, well, let's start on the top. What's the first one you're thinking?

I think margins, right? DTC margins are brutal right now. They've always been tough. And I think they've gotten worse. Customer acquisition costs are up, Ad costs are up, right? Shipping is up. Returns are up. You know, brands are looking at every line item and asking, do we need this?

Yeah. And the reality is retention teams are expensive. A retention manager, maybe $90,000 to $130,000. You add an email marketer, that's $70,000 to $90,000. A designer, for example, is like $80,000 to $100,000. And you stack all 4 people and you're spending $400,000 plus benefits Before you've sent a single email or SMS.

Right. And that doesn't account for tools, right? For example, Klaviyo is so dang expensive. And I think that's probably why people are cutting Klaviyo, moving to things like Omnisend, right? It's a lot cheaper. Um, and then, you know, if people had office space, that also is super expensive. And then any type of management that's overseeing the email team that's not actually on the email team, like, dude, it adds up fast.

Yeah, and I'll add one more. It's the hiring process, right? And turnover. I mean, those 2 things are just expensive. They take time. And it ultimately, at an operational level, if you think about a fully loaded cost of a retention team, you can get up to like $500,000, $600,000 when you add it up a year on just cost of overhead.

Crazy. And then if the brand's big enough, they have a CFO that's looking at this being like, what's the ROI? of this versus a contractor versus AI? And then dude, if it's a founder, they're looking at this like, shoot, this is money that could be in my own personal pocket.

Yeah, no, that's a fair question. And I would even add an agency is something that people are thinking about more and more, which shout out to the agencies out there. Obviously you guys are providing a service that helps cut overhead ultimately often, right? And you guys have your own teams. But you know, it used to be a hard question to answer because the alternative was worse, but You couldn't just contract it out to an agency and get the same quality, but I would say more than ever now you can.

Yep. Yeah, I think the math is, it's brutal. It's hard out there. It's hard for employers, it's hard for employees. I don't know, it's not really a great win-win standpoint at the moment. You know, if you're a brand doing, let's call it $20 million and email's 30% of revenue, that's about $6 million from retention. But if your team costs $600,000 or whatever that number was, You know, you're spending 10% of email revenue on the team that produces it.

Yeah. And now let's imagine you're doing the same, like $6 million with one person making $120,000 and I don't know, $30,000 a year in tools. That's still $150K, but you know what I mean? Like you freed up, you'll free up like $450K out of that math that you just kind of shared. And that's enough to drop the bottom line or fund a whole new acquisition channel. And we all know, Brands love acquisition.

Right. And we're not saying that you fire people or that you need to do this. It's just what we're seeing. And, you know, maybe you go higher with the savings people on other teams where maybe you need them and you're lighter. So, right, every CFO right now unfortunately is doing this math, every single one. And yes, there are CFOs in DTC. I think a lot of people don't believe that, but I think the second force, I think this one is big. Is AI is getting good. Like it's getting really good by the day.

No, dude. You know, we were always joking, but a year ago, you know how images and AI copy used to be and it was kind of bad, right? And you'd spend more time fixing than just writing it yourself. And those were the early ChatGPT days.

Yeah. Oh yeah. Um, you know, every email started with, in today's fast-paced world, or are you tired of?

Oh dude, I love this one. I always remember this one because it used to spit at me and I would yell at it, but it would just say, imagine a world where blah, blah, blah, right? And I'd be like, goddammit, chat.

Oh my gosh, dude. The whole templaty AI voice, right? You could spot it from a mile away. It was easy.

Yeah, but now as we've been talking about on the last couple pods, with the right prompts and the right brand voice context and the right workflows, well, you can produce a campaign brief, 3 subject lines, full body copy, an SMS variation, probably in like 20, 25 minutes.

Yeah, I agree with you. And I think the big part that people are missing is the quality bar isn't perfect, right? No, it's not. That's okay. But it's better than what most overworked folks are shipping, right? Last minute at 4 o'clock on a Friday, at midnight, whatever that looks like. It's Chase here. I hope you're enjoying the pod, but I want to get just 30 seconds of your time real quick to tell you about one of our sponsors who let us make this great podcast. So first I want to start with a question for you. Do you ever find yourself staring at a blank page, wondering what to write? Page waiting for inspiration to strike? Your marketing campaigns don't have to suffer when your creativity runs dry. Omnisend has recently launched their most powerful AI update yet, and it is awesome. Its AI completely transforms how you can create and send your next marketing messages. And here's really what's incredible. Their AI doesn't just write your emails, but it actually helps you write really compelling subject lines, preheader text that are geared to increase your open rates and ultimately build customer segments in a heartbeat and just whisper to the AI who you're targeting, and it's gonna figure out all the complex segmentation for you. That way you have no more guessing games. But hey, wait, it gets better. Their AI writer actually now proactively suggests improvements to your copy, and it's gonna adapt to your brand's unique voice and tone while following email marketing best practices. And their personalized product reminders use your browsing behavior to serve up exactly what each of your customers wants and needs to see. And their campaign booster automatically resends emails to people who didn't open or engage with them, which helps lift revenue by an average of 13%. And now with over 150,000 very happy customers already dominating their markets using Omnisend's email, SMS, and push notifications. So stop wrestling with writer's block and start letting AI do the heavy lifting for you. Visit omnisend.com and experience the future of marketing automation. Let them know that Jimmy and I sent you. And now back to the pod.

Those are some hard, truthful words there.

Yeah. And then let's be honest, like about what most kind of retention marketers' work looks like. So you've got a coordinator who's been at the brand for 9 months. They're juggling 30 sends a month. 3 flows are broken. You know, your designer is way backed up and they're trying to ship a campaign on Tuesday that the merchandising team gave 'em last minute on Monday night. Right. It's just tough.

Yeah, that sounds about right. And the copy that comes outta that situation, that ain't world-class either.

That's right. And the bar isn't AI versus the best copywriter in the world. I think the new bar is AI plus a smart operator versus a tired team running an ungodly number of campaigns a month under immense stress and pressure.

Yeah. And then you take that with comparison with what AI is doing these days and you have an AI with a smart operator or a prompter, they'll win every time almost, especially with consistency these days.

Yeah, I think that's important. AI doesn't really have bad days, fortunately or unfortunately, right? It doesn't have hangovers, doesn't get into a fight with their partner before Monday morning. It doesn't disagree with the way in which you want to do things. It just does it. It just produces.

Yeah, except for the occasional downtime from AI that we do have, but that's usually only a blip for a few minutes. I just had one this morning. But Chase, what was the third one you were mentioning there? What was the third force you think about?

The, the tooling. So Omnisend being our kind of primary and our favorite, Klaviyo, who's the more expensive Omnisend, Postscript, Attentive, right? They're all baked AI into the product now. Plus you also have third-party tools doing everything from creative to segmentation to send time optimization, deliverability monitoring. There's a ton of shit.

Yeah, I mean, dude, there is some cool tools that we saw and we're talking about some of them like Allen, which is a really cool tool, which is basically It allows you to load a ton of creative and then like it takes and makes a personalized email one-on-one for everybody. Or you've got Outcraft, for example. That's another cool tool. That one is doing like AI voice and abandoned carts and things like that. Like again, replacing the human, adding more incremental revenue for basically paying for what is actually driving money for your bottom line, which has been really cool. And the coolest part to me about all of this is that You don't need specialists anymore. You don't need an Omnisend specialist or a Klaviyo specialist or a Postscript specialist to run the thing anymore. They've gotten so good and easy.

Right. You know, 5 years ago, the send time stuff was deliverability consultants, like full-time job, dude. You'd hire someone for $150, $200 an hour, whatever it was, to look at engagement patterns, and then they would tell you when to press send.

Yeah. And now it's just like a switch in the ESP. It's that easy.

Crazy. And then if you look at things like predictive segmentation, right? That used to require a data analyst. Like you need someone with a PhD to just help you out, right? But now again, to your point, it's a toggle. A/B testing lines, subject lines, right? Like the same thing is now a button where it used to be a copywriter spending hours and hours on iterations and variations and it was exhausting.

And even design stuff, look, it's not perfect. But it's getting better and better in image generation, being on brand, layout optimization, assisting you with almost feedback and all that stuff. It's all there.

And it compounds across the stack, right? Every workflow that used to take, you know, human is now fully automated or AI-assisted or human has superpowers.

Yeah, I would agree. So look, when you stack these 3 forces, I think it creates margin pressure. There's AI quality, there's tooling maturity. You get like almost a perfect storm. And well, this is why brands don't need that 5-person team anymore. And frankly, they can't afford it anymore, which is why it's happening.

Yeah, man, it's hard. And I think it leads us unfortunately to where we are at now, where I kind of come around to like the whole 1-person thing that you mentioned at the beginning. Doing more with less because that's kind of the way of the world, right? Yeah.

And the wild part is, I'll say, is that the ones that are successfully kind of minimizing those teams, they tend to be happier doing it that way.

Why do you think that is?

Because, you know, the reason why you had so many people is because of like the grunt work, right? And they're doing the strategic work, which is like the fun work in my opinion. You and I both like to spend on that time. You tell us to design the email or create the email, that sucks. But what is fun is like, Strategizing a marketing plan and thinking about like what you can actually create and the work that actually goes into what the marketing is for.

Right. I think that's a great point. Most retention marketers I know didn't get into this to push pixels and copy paste in Omnisend, right? They got into it because they like the strategy of customer behavior. It's interesting.

And yeah, and now that's their sole focus and they just direct AI to do things. All right, well, let's move on from this conversation because I think we're starting to kind of beat on it a little bit, but let's get a little bit more tactical because if you're a one-person team or a small team, what does your job actually look like in 2026? What do you think that it looks like?

Yeah, I think the biggest change is how people actually spend their time. So I think we kind of alluded to this, but like the old retention manager, dude, they could spend 70% of their time just making things, writing copy, building emails, designing assets, and only a small portion, I guess, what, 30% or so on strategy, where now, you know, it kind of flips in a sense where you should spend a bulk majority, maybe that's the 70% now on strategy and the kind of remaining percentage of that 30% is on production where AI is gonna handle the production grunt work that you never wanted to do in the first place. So, you know, the, Actual job of one person kind of comes down to, I guess, 4 things, right? The first is you now as the human, you own the strategy. What are we sending? Why are we sending this? Who are we sending this to? And when, right? Think about like the campaign calendar, lifecycle map, and frankly, kind of like the segmentation logic.

Yeah. And this is, and let me add to that, this is like the highest leverage point Of the job anyways, and the part AI can't really do yet successfully. You still need someone who understands the brand, the customer, the product, the margins, and really the seasonality.

Yep. Yeah, 100%. And, you know, AI can suggest a calendar, but it can't tell you that your founder is launching a new product in March or April that you need to seed in February with educational content. They need you, right? That's the human knowledge.

Yeah, I agree with you there. And again, most retention marketers also were never doing this part anyways, or they weren't doing it well because they're generally buried in production work and they always were a little bit more reactive, I would say. And that's definitely changing right now.

Yeah, right. The calendar used to get built a day, maybe a week, 2 weeks max, and maybe only for a snapshot of a few weeks. Because that's all the headspace that people would have. And it makes sense, rightfully so.

Yeah. And now they got like more time to think about it.

Right. And I think the brands that empower their retention folks, their retention person, I guess, in this kind of perspective to be strategic, are going to see way better and more immediate results, I think.

Yeah. And you know that skincare brand that I talked about earlier, you know, the retention manager was great but completely buried in work. Basically, right? And I helped them find different AI tooling, automations in different ways, like things that we can do to like really reduce production work. And that reduced it about 60%, really. And within 90 days, she was able to remap the new lifecycle, segment their lists 5 different ways, and launch 3 new flows. Pretty big work.

I love that. 3 new flows is just— that's wild.

Yeah. You know, what it was is like she just didn't have time in the past to build those flows and the strategic work that she wanted to do was sitting right there, but she just literally didn't have the bandwidth because as you said, it was reactive, right? Like she was always fighting to catch up.

A quick break to talk about Super Scale. If you've ever found yourself on a Friday afternoon scrolling through the Meta ad library, screenshotting your competitors' hooks and trying to reverse engineer why their ads are working while yours personally have been stalled, that's exactly the problem that SuperScale is built to solve. It's the first AI agent specifically for social marketing. You connect it to your Meta, TikTok, and your Google ad accounts, point it at your site, add your ICP, and pass testing notes, and it actually goes and does the work. How cool is that, right? You can literally prompt it to analyze your top competitors, identify what's converting in your space, and turn that into a new batch of creatives across different formats, all from a single prompt. What would normally take a designer and a growth lead a few days can now be done in a matter of minutes. They're backed by Cradium and trusted by brands like Taxfix and SumUp. You can check them out at superscale.ai. Now to the show. Yeah, it's, I think that's big then. I think it's big for the company. I think it's big for the person actually doing it. Um, I think most retention programs are probably 60% of what they could be because the team frankly doesn't have time to build the other 40%. All right, so now let's move on to the second job. You are now the director. You're directing the AI. You should not be writing the copy yourself. You should be writing the brief that gets AI to produce the right copy.

Dude, that's the real skill when you think about it. People underestimate it, but if you're not learning this and you're listening to this pod, like, this is where you should be spending a lot of time because a good operator can get you usable op Output within a couple prompts, right? And a bad one, well, they're still tweaking after 20 prompts.

For sure. And, and there is a learning curve, right? The first couple weeks, the first month you're going to be slow, you know, then the next month you're starting to slowly figure it out. And by month 3, dude, you're reaping the rewards. You're flying.

Yeah, that's definitely the right thing. And I'll put something out there for people listening to this pod, a really good thing. You know, you should be building kind of reusable prompt templates and saving them for your brand. It's one of the most highest ROI things you can do. Just have an Excel and use Just dump your prompts that have been working and the things that you're getting. Because if you have a brand voice prompt, a campaign prompt, SMS prompt, an abandoned flow prompt, all saved, all loaded with context, it just saves so much time and the learning curve just gets so much lower.

Yeah.

And, you know, we essentially have like our own kind of prompt library that we share with clients and the team, et cetera. Where, you know, brand voice gets kind of dialed in. And then after that, every campaign brief is just plugged into the chat for the specific project client, you know, KPI, whatever it is that we're doing. And it's just adding minor details. It's pretty great.

Yeah, I mean, it's for everything we do, I think, within our businesses together and other businesses that we run. I think we all have these cool prompt libraries that we're building and I'm probably like 40, 50 prompts deep at this point. And every time I find a new angle that works, something that works really well for me, well, I save it so I can use it again.

Yeah. And I find that like the compounding on prompts is actually like, you know, very, very underrated where every good prompt then makes, you know, the kind of the sequential ones even better.

Yeah. So look, if you're listening to this still and you're thinking about this and you're not already building A prompt library. This is your call to action to say start building one today because each, even just like a single campaign brief prompt, that alone could save you hours of a week basically. And it could save you a lot of headache into the future.

100%. I think like the third job then would probably have to be like QC, right? Focusing on quality control where you really, as the human, you're the filter. AI is going to get you to, let's call it 80%, and you are going to really focus on that core 20% that gets it over the finish line to 100%.

I like that quality control because this is where most people screw up. They don't even edit. I see it too often and I see it a lot actually. They just, they don't edit. They ship that robotic AI copy or they over-edit and waste all the time savings. I think I've seen both sides of the spectrum.

Yeah. And I think both ends of the spectrum are frankly wrong, right? If you either ship pure kind of AI copy, then you sound like every other brand. And if you over-edit, um, you kind of lose the benefit of the speed.

Yeah. And this is where the prompting skill and just your general skill of knowing what, what's worth editing and what's not, like subject lines, you're always gonna have to edit. Those aren't gonna be great. Right. Opening lines, that's the hook, man. That's so important. You should be editing. A body copy, once you start getting into it and you've got the reader's attention, that doesn't require a lot of editing. And like a footer, man, like God, you don't need to spend any time on that. That's whatever at that point.

Yeah. And I'd add like the close, the kind of the call to action, the CTA. I think that's where a lot of the conversion actually does happen. So I do believe that's a really core piece kind of to edit.

Yeah. Okay. Yeah, I agree with that. So subject, hook, close, those are your 3 non-negotiable edits. Everything else, it's kind of contextual.

Yeah.

Yeah, 100%. I, I think hopefully that this is kind of helpful for people. And at the end of the day, like the brand voice is the moat. AI can't read your founder's mind. What you actually can.

Yeah. And this is something I want to drive home real quick as we finish this one off. But each brand has like different ticks, right? Words they use, words that they avoid, phrases that feel right, phrases that feel off-brand. That stuff doesn't come naturally to AI unless You tell it and you train it.

Yeah. All right. So should we kind of move on, I guess, then to, you know, really talking about like when you train it and kind of needing a human to catch the stuff? Like, should we kind of keep going on this angle?

Yeah. I mean, look, we don't have to keep going on it, but I think, you know, what I would like to think about is like, okay, so here's a perfect example. That beauty brand that I was talking about, they use the word ritual instead of routine. Like, that was like a small thing that they did, right? AI kept defaulting to routine no matter how much you told it because that was a more common word. And often we just keep working on it. We eventually figured out how to get the prompt to figure out to use the consistent word ritual instead of routine. And so like these are things you have to work on.

Yeah, I like that. And like those kind of like small details, that really is what kind of I think distinguishes a brand. Like you either care about it or you don't.

Yeah. And look, I'll add one last point. Let's move to the 4th one. But if you don't care, your customers can tell. Make it, make, maybe not consciously, but they can tell and they know that this is not the, the voice that they're used to hearing. So anyways, let's go to the 4th one, Chase.

Cool. Yeah, I think the 4th one comes down to looking at the numbers. So what's working, what's not, and what to test next.

This is a good one, man, because we just had a, depending on when you're listening to this pod, we had a pod right before this one where we talked about this stuff. But AI's helping this massively. You can dump your CSV of your last 30 sends in the Claude or ChatGPT, or you could connect an MCP to it. And ask it to find patterns, subject lines, send times, sub-segment performance. It can give you so much information.

Yeah.

And I think this used to take like multiple and multiple hours, dude. I feel like sometimes I was banging my head against the wall on a 4-hour session analysis, but now I wanna say it's taking the team in the ballpark of like 15 to 20-ish minutes. Yeah.

And even now I think that's slow considering what we'll get to over the next like couple years here. But right, I think with the analysis getting better, it's, And what I like is that it's not biased. You have your favorite campaign that you want to defend, but AI, they don't care. They're looking at the data and they want to tell you what the data says.

Yeah, I think it's a real thing. I think marketers get too attached to their work where AI has no ego. AI doesn't. Yeah.

I mean, perfect example was like, and I fall into this category too just as much because we're always proud of our work, but I had a campaign that I thought was a winner because the creative was great. It felt great. It read great. The hook felt great. And I dumped that data into Claude and asked it for patterns basically. And it turned out the campaign itself actually kind of underperformed because the segment wasn't that great, had nothing to do with the creative.

Interesting. And that wasn't something that you necessarily maybe would've caught yourself, right?

No, man. I would've never caught it. I would've never seen the segment there. I was just too close to it.

Yeah. So I think that's, I think that's spot on. And you know, those 4 jobs that we talked about, I think the first was strategy, AI direction, quality control, and then analysis. That frankly is the entire job. And I think 1, maybe 2 good operators can do all of it for a brand doing 7, 8, potentially 9 figures, but at least 7, 8 figures for sure.

Yeah, no, I agree with you and I've seen it. I've seen it in the market. It's happening more and more and operators who are good at this are making More, they're making more money than their old retention managers used to make because they're producing more value ultimately.

And I think that's important, right? So I think jobs unfortunately do consolidate, but I think the comp is going to go up for people who can do it. So can the people that aren't needed on the existing teams go find companies to do it that don't have this? I think, I don't think people are getting fired, but I think people are going to be shuffled.

Yeah.

Oh dude, no doubt. And you know, I've been watching the comps because look, salaries are going up, but I will say the email marketing or the retention manager role is going up in dollars. I've seen roles at $150K, $200K easily, and that's really big compared to back in the day when, you know, $75K used to be the number. So there is more, but the brands are asking for more work, which is fair, right? And they're replacing this $600K headcount like we talked about before down to one person banking $150K, $200K, and ultimately, Often making more money for the brand.

For sure. And I think unfortunately people are going to listen to this and just be super worried. Um, I would say, don't worry. Um, the path is clear. You just need to become the person that can do this and there's going to be job security. I mean, dude, to make $150K to $200K doing this, um, I think it's gonna be great because you just now got to focus on the fun. You get to work with sick brands and you don't have to do a ton of extra stuff. Like, I think it's going to be awesome.

Yeah. Okay. Let's shift gears here. Let's talk about the stack. What does the AI stack look like for a one-person retention team in your opinion when you think about it?

Yeah, that's a good question. I ultimately think it comes down to kind of 4 different layers. I'll start with the first. Layer 1 is going to be kind of like your general purpose model, for lack of a better word. That'd be things like Claude or ChatGPT. And this is going to be like the daily driver. Where you're writing copy, where you're submitting briefs, where you're running analysis, and kind of like your brainstorming partner.

Okay, I can see that. And you're using it probably with some custom instructions and projects loaded with your brand voice, your customer profile, your past campaigns, and not just cold prompts, right?

That's correct.

The difference between like the cold prompt and the brand-loaded project and the prompt that's ready to rock and roll, Dude, you know, there's a tier of operator out there who opens ChatGPT and types in, write me an email for a sale. And look, that's never going to produce good work. And there's a tier of operator who got the project loaded with a brand voice doc and an ideal customer document, 20 last campaigns. They've maybe connected their ESP like Omnisend to their MCP, brand guidelines and tone examples. And that person, they're going to be able to produce good work. When, when they prompt, they get gold every time. That's the 2 sides of the story that I see right now.

Yeah, I'm with you. And it's actually not that hard to set up. You know, I think to do like a good brand context project, it takes maybe a couple hours to build out the first time. And then from there, like every campaign benefits from it, you know, for, for forever, frankly.

Yeah. And I'll add real quick, this is a tip for everyone listening. You should have projects. You should be using projects for every brand. that you work with, anyone you're working with. If you're an agency, you've got multiple, or you're a single brand, you should be loading into a project and that should have your brand voice doc, your ICP or your ideal customer doc, your like top 20 past emails that you've identified, any brand guidelines, and then you start from there and that will definitely work. Yes.

Yeah. And then, you know, I'd add to like the top 20 past emails piece is I think so huge. You know, that's It's kind of like your style guide more than the actual style guide. The AI is going to be able to see how the brand actually writes, not how you actually claim to write.

Mm-hmm. Great point there. 'Cause the brand voice doc is what you say you do. The past emails is what you kind of actually do, and AI learns better from the actual stuff.

Right. And then moving on to the next layer. So layer 2 is really focused on like research inspiration, and this is where, uh, inbox.ai, which we built, fits in for For this perfectly. You know, it pulls what's working in your category. You can see how your competitors are positioning, and it's going to allow you and the AI to study the best top-performing brands and emails.

Yeah, you definitely don't want to start with a blank page. I think that those days are gone. You want to start with context. The more context, the better. And what great way then to learn from other brands in your industry.

Yeah. And you actually want to start with what's being sent today, not multi-year-old swipe files.

Yeah, this is something I think that continues to get overlooked. The category context definitely matters. The skincare email, a supplement email, a coffee email, they're all different rhythms, objections, and different buying cycles. So you've got to kind of cater to what's working in that market.

Yeah, and you, you can't just generate emails like in a vacuum. You really need to understand what your customer's seeing in their inbox from competitors and just other brands in the market.

Yeah. Okay. So if everyone in your category is running the same limited time angle, you need to know that you can be different or you can run it harder than they are, but you can't make that decision blind, which is why we love inbooks.ai. So layer 3 here, I would add to here is production. I think this is huge because look, your ESP these days have built-in AI layout, image generation tools for hero design, AI. Or on-brand visuals, or you can use a tool like Allen. That's another great tool that does this too, right? It layers on top of your ESP. But you mentioned this before, Klaviyo, Omnisend, they all have AI features now and subject line testing, send time testing, predictive segmentation. It's all kind of there. Wouldn't you agree?

For sure. And I think the image side is getting better. You can generate on-brand product lifestyle imagery in minutes where it used to take You know, non-designers like myself, weeks if ever. And then it used to be designers like full week of work just being bottled down in this stuff.

Yeah, but I'll add one thing. And this is at a consumer eye that I've noticed is that AI imagery sometimes, especially like product shots, you can almost tell that they're not real, like use it for back— like using AI for like background lifestyle, putting a shirt on a model, Or a coffee cup in an abstract area, whatever it might be that works. But just the product itself, I still think it takes a little bit of that human eye.

Yeah, I'm with you. And don't use it for people if you can avoid it. I think AI people, AI people still have uncanny valley things. Hands are weird, eyes are off. Something just feels wrong about it.

I see it on YouTube all the time. I see ads where it's like, I'll watch it and be like, that's totally AI generated. I know it is just right, literally from watching their movement and behavior and the way they talk and stuff like that. So yeah, real models, real product shots. I think AI can help transform it. AI generated backgrounds, design elements I think are cool. I think that's the formula. And you know, look, on the design side where I see most underutilization right now is most operators are still building emails the old way. They're using the old drag and drop. Manual everything. There's tools that'll generate these on-brands based on the design system in just minutes, or you can load them in and have them flip them around. Again, plugging Alan here, but that's a great tool that you can look at. We'll make sure we put the link in the descriptions below.

Yeah, which when you do that, right, like it just gives you more variation, which means more testing and better results.

Yep.

All right, so layer 4, let's talk about that one.

Yeah, I'd say this one is focused on analytics. So AI is digging into your data, ChatGPT with code interpreter, Claude with file uploads, or like an MCP that connects to your ESP, like Omnisend, their MCP is pretty sweet.

Yeah, I mean, we just did that pod 1 or 2 ago. So after you listen to this one, jump over that one if you haven't heard it, but I highly recommend it. And, you know, this is a layer that most people are skipping right now. They use AI to make up, make stuff, but not to learn from what they made. That's probably the way I think about it.

Well, I like, I like that. And I think that's where like the, the growth and the compounding wins actually begin and start to happen is every campaign that you send becomes training data for the next one. And if you're not analyzing it, you're just spinning your wheels.

Yeah. Let me give you an example of a great workflow for those that are listening. So every Monday, basically, I dump last week's data into Claude. And I ask it 3 questions. What are my top 3 performing and what did they have in common? What were my bottom 3 and what went wrong? What should we test this week based on those patterns? Like those are the kind of things that I ask it and it gives me all the information.

Dude, that's money. Um, right? Like that used to be like a 3-hour analyst report and then you'd have to like try to understand what they're saying. That now is like really kind of quick back and forth, almost like a meeting that you're having with your operating partner being the AI. And it's just way more efficient.

Yeah. And I'll add to that, AI catches all the stuff you miss, patterns across send times and subject line lengths and emoji usage and segment behavior, things that are really hard to look at without really digging in.

Yeah, I had recently kind of an AI analysis that surfaced kind of one of my clients' segments. So for example, I think it was like their VIP segment. It was surprisingly performing worse than their general list on a specific type of campaign. And that was for sales. Like the VIP, for whatever reason, just ignored sales because they had bought it at full price, right? They didn't really value the discount or the sale. They just wanted exclusivity, which I thought was kind of an interesting finding.

Dude, that's a great way to explain. Basically, without analysis, you would've just kept blasting them with the discounts, right?

Yeah. I mean, look, the data's been there the whole time. We just hadn't asked the right questions to uncover what was going on.

Yep. That's the magic right there with AI. It doesn't have the insights that you don't already have. So don't think of it like that. It's really, it asks better questions of your data.

Correct. Or ask the questions you didn't have time to ask in the first place.

Yeah. Same thing in the end, at the end of the day. So anyways, let's go back to the layer of the 4.

Sure. So the kind of the general AI, the research inspiration, the production tools, and the analytics. All need to be loaded with brand context and all need to be working together.

Okay. So those 4 layers stack that an operator has, this is basically going to run circles around a team of 5 basically. That's what we're kind of saying as we look into the future here. So Chase, let me play devil's advocate because I think that people always want to hear this. What breaks when you have a 1-person team? Because it's not all upside. I don't want everyone to go off and replace their teams right now. I want to make sure that they have a good understanding of this stuff. What's the downside here? Hey, you guys want to stay connected with us? Guess what? We send a daily retention newsletter out that you should join right now. If you're not a part of it, it's at ecomemailmarketer.com. It's completely free. We send it to your inbox daily. We do breakdowns. We talk about the latest things. We talk about the latest tools out there. There's so much you can do. Plus we even talk about the pod, of course. So if you're enjoying the pod and you want to be here, you want to listen to us, you want more email and SMS retention marketing, Tips, tactics, tools, tricks, whatever it might be. Head over to ecomemailmarketer.com. We probably have something free that we're giving away right now too on top of that. So go get that, download it, join our newsletter. We'll see you there.

Yeah, look, I mean, there are real, real trade-offs. So please don't go fire your team. Please don't quit your job because you think you're going to get laid off. Like, you're not. Like, you're going to be okay. But we do need to be honest about kind of what the trade-offs are, because if you do go into this thinking it's all wins, Um, it's going to be very, very easy to get burned. I think like the biggest risk here is like the single point of failure. So if the person that's running the show gets sick, they take vacation, if they get poached, dude, your entire program is on pause. It goes down.

Oh man, I didn't— that scares me to even think about it. And retention is recurring revenue. If you send slip for like 2 weeks, well, you're going to probably feel it in the cash flow. Wouldn't you agree?

Dude, yeah, immediately. If you miss sends for 2 weeks, um, in a healthy retention program, like an 8-figure brand, that's 6 figures of revenue that's just gone. Sometimes more that just, it evaporates. It vanishes.

Yeah. So solution time, because I don't like talking about problems without some type of solution. Uh, I would tell people documentation becomes important. I know it's a big corporate world, but the reality is like every workflow, every prompt, Every SOP you should be writing down, you should be recording things so that the next person you should be able to onboard quicker, not in 3 months. Like that is the biggest thing I would tell people who are listening to this.

Yeah, I think that's a really good point. And I think the documentation makes that one person better at their job anyways, because it allows them and forces them to actually focus on clarity.

Yeah. And dude, we do this with our clients right now, right? We give them a document where we want to document everything. We want to understand where things are. Things are the point of failure. And then the reason why we do it is because we don't want it to be all on one person. We want this information to be able to be spread so that if something does happen, that information can be used by someone else to pick up the work and continue on with the things that we need to do.

Yeah. Plus, and this is something I think that is easy to miss, is when the person does take a vacation, because they're going to need to, who covers there, dude? Like you do need a backup plan. You need a fractional contractor on retainer, trusted agency partner. Someone needs to do something at all points in time.

Oh yeah. This is a good one because I actually know someone. I actually know someone who does this. She runs a retention for like 2 brands solo, and then she's got a contractor on retainer who basically steps in for like 2 weeks if she needs to disappear, like go on vacation, do stuff. And it costs her like $2,000, $3,000 a month for insurance. I know it's a little bit expensive, but again, You're owning more, you're getting paid more. That way you can afford more.

That's really interesting. I've, I've actually never heard about that. Um, and I don't think most people think about this until they need it and then it's too late. Okay. Um, I think like another risk I was thinking about as we were talking would be quality drift. So if one person's doing everything, dude, they get tired. I'm tired just thinking about that. And they're gonna cut corners. Um, and AI output starts shipping. With fewer and fewer edits being made.

Yeah. And as I mentioned before, the customer will notice this. And before the marketer does, the open rates start sliding, click rates start dipping, and you start seeing more unsubscribes. I mean, and these are things that you'll end up seeing because your quality's not there.

That's right. And by the time that you noticed, you know, I won't want to say it's too late, but you've already lost a lot of ground and footing. You know, your reputation with inbox providers takes a hit, as does your engagement. Kind of in the wrong way.

Yeah. And this is why you need a quality bar or QA, basically, that it's enforced even when you're tired. A checklist, a subject line passes that test, a preview text is set, mobile view is checked, like links are tested. Like you still need an old school traditional checklist to make sure all the things are happening, right?

I know it's boring as heck, dude, but it is so necessary.

Yeah. And one last thing I'll add to that is schedule audits because like real ones every 90 days, somebody else should look at your work. If you're a founder, you should be looking at that work. It could be a peer, it could be a fractional advisor, it could be someone you bring in to audit, or it could be someone in a community you're in, for example. But somebody needs to be looking at the sends with fresh eyes because those are the people that are going to help you spot things really fast.

Yeah. I think that makes a lot of sense, especially because you often go blind to your own patterns. Like, you don't know what you don't know.

Yeah, I'm guilty of that too. And every time I see that, you know, the— you ship the same hero image style for 6 months without realizing it. We've seen it within ourselves. We've seen it with some of the clients we work because honestly, nobody's telling them and you need somebody to tell them that. Chase, what else we got for risks? Let's keep going.

I think another risk that comes to mind is like tunnel vision. You know, if you're alone, you don't have anyone to push back on your ideas. And your personal weaknesses and blind spots become therefore the brand's weaknesses.

Yeah, that's a big one, man. A team has natural friction, one brand or one person, they don't have anything.

Yeah, yeah, I think friction is good. You don't always feel like it in the moment. But I actually think in the long run, Some degree of friction is important.

Yeah. All right. So let me tell you a story of one. I had a client where a retention manager was so good at her job that everyone kind of deferred to her. Every strategy decision was hers. And for 2 years, there was like no pushback.

And what'd that lead to? What happened?

Well, and she had this fixed idea about her cadence, right? She believed strongly that they should send 3 times a week max, no more. It was working. So no one, nobody questioned it ever.

Hmm. Until you came in?

Yeah. Until I came in and asked the question, have we tested 4 sends a week? 5? She said no because she knew 3 was the right number.

Interesting. So how did she know?

She didn't.

That was the answer. She didn't know if she did. So we tested it and we went to 5 sends a week and it beat it by 40% of revenue to 3 with basically no impact on subs, basically.

I think you said 2 years, dude. So 2 years of just unfortunately leaving money on the table because no one had the guts to push back.

Yep. 2 years. And she was a great operator and just had blind spots. And there was no fault to hers. Everyone has them. I have them. But when you're solo and you don't have people challenging you, well, that's just the output that can happen in this market. So this is a downside, 100%. I agree with you. That could happen.

Yeah, it's a, that's a good one. Um, I think the, the solution is external, right? Like you get into the mastermind, join a community, hire a fractional advisor for a month, you know, at a time just to have someone to push back on your strategy.

Yeah. Or have AI do it, right? Drop your strategy doc in the cloud and ask it to argue with you. You know, one of my favorite things that I do is I'll have 2 AIs fact-check each other. I'll have something on one AI do it and I'll throw it in another AI and tell it to rip it apart. And believe it or not, rip it apart will fix it. And I'll throw it back to the other AI and say, I fixed it. I ripped it apart. What do you think? And it will rip it apart, right? Like you can have it come, utilize it by basically utilizing the ones that you have available. And there's way more AIs, Gemini, Perplexity, Grok, like all these AIs out there right now that you could be using. Have you ever done that?

Yeah, I do do that. I don't necessarily go from like one AI to the next to the next and kind of, you know, pin them against each other. But I do often in the chats like push pretty hard on like, what would a contrarian say about this approach? Or what are 3 reasons this might fail? Or, you know, what are we missing here?

Yeah.

And I think most people would be surprised how often it catches its own stuff that it's doing. All right. 4th risk. Let's talk about one more risk. What do you got?

I think the creative sameness is kind of the 4th one.

Okay, I like that one. Talk— let's talk more about it.

So essentially, like, when everyone is using the same AI tools with similar prompts, every brand starts to sound the same, and then inboxes are actually flooded with the same AI-flavored copy.

Yeah, dude, I've been able to spot AI emails just by reading them at this point, or even social posts or anything like that. Certain rhythms, certain phrases, certain structures, like, they're all right there. And once you see it, You just can't unsee it.

Right. And the customers see it too. Not always consciously, but there's a flatness to a lot of these emails now that people just feel.

Yeah.

And look, the brands that are winning in 2026, the ones that figure out how to use AI for speed but keep their voice distinct, their brand on voice, they come back to the brand voice doc, your actual founder doc, all that stuff in your customer doc. That's all things that people need.

100, 100%. I think at the end of the day, like AI is a great accelerator, but it's not the best originator.

Okay, so you know, I would say what use it for 80%. I think that's what we keep saying. Bring in the 20%. That's going to be yourself. That's the biggest difference between the brands that sound like everyone else and the brands that sound like itself.

And that 20%, right? Mostly the personality stuff. It's opinions, the takes, the little sides. That stuff makes someone go like, oh yeah, this is from X, Y, and Z brand. That makes sense.

I like that. AI can't do that yet, but not reliably, I would say. So this is where you spend your time editing, and it's important that you're putting that human effort into it. Hey, I want to add a 5th one, Chase, and this is kind of a hot one, but burnout. I think that's one that I think about.

Yeah, I think that's a great one.

Yeah, dude, when you're working alone and you're working by yourself, the work never feels done. There's always another campaign, another flow to optimize, another segment to test. And without a teammate, there's like nobody to share the load and it gets really heavy sometimes, I would say.

Yeah. Plus you've got no one to celebrate the wins with.

Yeah. The emotional side is real and operators who go solo or don't have a community around them or anyone to talk to, they burn out fast and I've definitely seen it.

What do you feel like is a good fix for this?

Man, it's no different than the tunnel vision one. I think an external community. I think we have a community you can join, for example, and, you know, you can come talk to other peers out there or a peer group of people locally that you like to, somewhere to vent to and trade ideas. This is really important. Friends, right? Basically, your role can be lonely if you let it be, in my opinion.

Yeah. Confide in someone, whether it be a therapist, a partner, I don't know, someone, right? Solo operators in these high-pressure roles, they do need somewhere to kind of process that stress or they hold onto it and it's hard.

Yeah, I think that's going to be a trend that we see where mental health is just part of the job description in 2026. You're going to have a one-person retention team. Well, you're going to have to take care of that person and make sure you, if you're a founder, an agency owner, you got to check in with your people. So just not to get too philosophical, but I wanted to kind of bring that up. All right, Chase, it's time to wrap this thing up because I think we've been going a little long. I love this topic. It's been good. So listen, if someone is listening to this, either an in-house marketer or brand owner, let's talk about a little playbook on how to operate as one-person army. I think we've kind of defined it. So you want to kind of kick this off? Let's do it in like 5 steps.

Sure. Yeah, 5 steps in order. So step 1 is really about getting your brand voice on paper, not vague stuff, but being very, very specific. Um, words you actually use, words you would never use, sentence structure, the tone, even examples of on-brand and off-brand copy.

Yeah. And this is like a one-time investment. I think you need to spend a day on it and it pays back forever, in my opinion.

Yeah, agreed. Agreed. Um, and when, when you say one day, right, you actually mean spend the day, right? Don't half-ass it. This document is gonna be the foundation for every prompt you write for The coming years and months.

Yeah, 100%. And, and also, uh, add negative examples. Here's what we don't wanna sound like. Don't do this. I think that's more important than what you sound like.

Yeah. Examples of competitor copy that's off-brand for you, enter AI-generated copy that misses the mark, old campaigns that didn't work. They're all super useful. And that leads us to step 2, which is building a prompt library. So you should have things like a campaign prompt, a flow prompt, An SMS prompt, a subject line prompt, an abandonment prompt. You need to save them and refine them over time.

Yeah, this is like your cookbook, I would say, right? Those same prompts you're going to use over and over again, and you're probably going to get a little sharper every time. And each version of them is going to be like a subject line prompt V1, V2, V3, right? So you can actually see the evolution and why. And that's a pro tip because most people don't version their prompts and they lose track of what worked ultimately. What's step 3 look like for you?

So this is where your calendar should be planned 90 days out minimum, you know, different types of themes, promotions, product launches, seasonal content. Um, and then once the calendar's actually launched, AI production becomes a fill-in-the-blank exercise because the strategic work is already done. And without the calendar, you're just being reactive. Um, AI won't save you from being reactive. It'll just make your reactive work faster.

I like that, which doesn't help really. Going faster is the wrong direction. You still, you're still going the wrong direction.

Yeah, that's right. And honestly, the 90-day calendar is what separates, I think, a good operator from a great one. The great ones know what they're sending 11 weeks down the road already, which then leads us to step 4. So every campaign goes through the same checklist before it goes out: subject line, preview text, copy, links, segment suppression, Mobile preview, et cetera.

Yeah. Uh, I would say this is what keeps the wheel on, right? It's boring, but it's critical.

Right. Um, and then the checklist should be physical, like on the wall or in a Notion doc that you actually open, not just in your head. 'Cause if it's in your head, you're gonna skip steps. Um, you're gonna make stuff up, you're gonna miss things. It needs to be a checklist where you don't skip steps. Um, and then step 5 is, this is the one that people skip, I think fairly often. Is you have to block time for strategic thinking, not just production, right? Not just analysis, but actually focusing on like the strategy and what's going to move the needle.

Yeah. 'Cause AI does 80%, right? If you don't, if you don't, the production work expands and to fill it with all your available time, you wanna make sure that AI is helping you as much as you can.

Yeah, I think that's correct. Um, I like to block out like a Monday morning type thing for The kind of the deep strategy work. I'm not really building emails here. I'm not sending campaigns. I'm frankly just thinking, planning, looking ahead. And this, like, you know, is part of what justifies the larger salaries that are happening. Anyone can press buttons, right? But the strategic thinking is what separates the operator from the AI. This is why you as the human are needed. And it's also probably the part that's hardest to defend. There's always going to be pressure to just ship a campaign, but if you give up strategy time, it's going to be very, very apparent down the road in six months. So you need to protect it because your job depends on it.

All right, cool. Well, quick recap to that: I guess the traditional retention team is getting restructured out of existence, in my humble opinion, because the margin pressure, AI quality, and tooling maturity. And what that's leading to is the one-person operator. job does 4 things: strategy, AI direction, quality control, and analysis. And AI handles the production for you. And your stack generally becomes this 4-layer stack, basically general-purpose AI, research and inspiration, of course, your production tools like Omnisend, and then of course analytics that you want all loaded with the brand context. What else you think?

Yeah, I think the risks are real here with like the single point of failure, the quality drift, tunnel vision. The creative safeness and the burnout. Like you need to plan for all of those because they are going to happen.

Yeah.

And a playbook, just to add the, make sure you're documenting stuff, right? Your brand voice, build a prompt library, lock your calendar for 90 days and do some kind of standard QC, some quality control and protect time for strategy. I think that's the most important thing in my mind.

Right.

And I think at the end of the day, this is really gonna allow teams to level out. It's not just about having the biggest team. It's about having the team that works the most efficiently with the right tool.

Yep. So listen, as we wrap this thing up, if you're a marketer listening to this thing thinking your job's at risk, it's not. It's you evolving and becoming an operator. That's the entire point of this pod here today is that you've got to go and start adapting and changing. The work isn't going away, it's just consolidating into fewer hands. But getting really good at directing AI, getting really good at strategy, getting really good at protecting your brand voice, That's a job for 2026 into the future. And you're going to get paid for it. These top operators, I've seen jobs at $150K to $200K plus, which is incredible these days for an email marketer or retention marketer because, well, they should. So with that being said, guys, this is a wrap to the Death of the Retention Team. I don't know what we're going to call this exactly, but if you enjoyed this, you got thoughts, you got questions, please DM us, comment below on YouTube if you're listening to it on YouTube. If you're on Spotify, jump over to YouTube and come check us out there. And if this episode hits home or it pissed you off, man, well, let us know. Or share it to someone and let us know your thoughts because we wanted to take a hot take here and I think we delivered with that. So thank you for listening. As always, links in the description for all the things we talked about and we might even have a downloadable. So go check that description out. Otherwise, we'll see you next week.

From the open to the click, they're the masters of the flow. Just watch your list grow and your leads turn to go. Did we entertain you so far, or do we have to extend it? If you want results, you know you gotta send it. Yeah, you know you gotta send it. Yeah, you know you gotta send it. Yeah, you know you gotta send it.