The Most Ingenious Way I've Seen Ad Performance Measured In Years (With Michael Ting From Jaxxon) transcript

Last week, I had lunch with Michael Ting, and I am telling you, this is like one of the smartest dudes in e-commerce I've ever come across. He's got a serious data and quantitative uh analysis background, and he brings that to Jackson, a large and growing men's jewelry and accessories business. And I'm just telling you, this conversation blew my mind in multiple ways. Uh when I got introduced to Michael, it was with basically the same kind of introduction that I'm giving you right now, which is like, "Dude, you have to talk to this guy. he's extremely smart and you've never met him before.

Turns out he lives in my hometown as well or where I live now anyway. And uh and so we were able to get lunch and it was a great conversation. We're bringing that conversation here onto the podcast for you. Now, I'm just telling you, you're going to hear things I have never heard anybody else say about how their marketing mix is optimized at uh at Jackson and how Michael thinks about that as the head of revenue at Jackson.

I'm not going to delay it anymore. Let's get into it with Michael Ting, head of revenue at Jackson. [music] Michael, thanks for taking Dude, you you already you already spent time with me at lunch. I I actually got in the car and realized we had spent longer than I like expected. So, uh so thanks thanks for hanging out with me at lunch. Thanks for hanging out with me again on the podcast. >> Yeah, definitely. It's pretty crazy that we were one one two miles away almost. >> I [laughter] know. >> So, >> yeah. our our mutual friend at Intelliggeems um Alex McKarn he uh he introduced us and um and he was like dude you've got to talk to this guy and then it's like I like pulled up your LinkedIn and it's like Torrance and I was like Torrance like it's like not a lot of us here I mean there's some there's a few few e-commerce people here uh but uh uh but yeah it was it's fun to do um hey I want to get into your background and credentials in a little bit but I actually want to get straight to the juiciest thing uh because I think people are going to be interested in Um, you said something to me I've literally never heard anybody say before that I've always kind of wondered about in the back of my head but not enough and I'm I'm probably just not a good enough data person to go do the math here.

But you told me that you guys your core metric I and I I'll say it the way I think you said it and then you can correct me if you need to. But um your core metric that you guys have sort of optimized for at Jackson and your marketing mix is cost per lead where where lead is an email address. um out of your CRM. Uh is that right? >> Yeah, it's actually super interesting. I have a good friend at Gruns um who was there very early on and is now their president and it was kind of inspired when I saw these supplement companies and hearing how they manage their advertising spend on like an LTB over CAC.

The problem is we're not LTB, we're not a subscription business. So it was like at what point can I view how can I pull forward when I attribute spend to a revenue? And what we found is that the advantage of a lead and email is that you can perfectly attribute the sale to the email and you can attribute the spend to when you got the email. And the combination of those two things just have made a really powerful metric for us.

And it's just about finding these key moments where marketing costs are low, buying intent is high. You're not seeing the sales yet, but you're seeing the interest. And that's really what it's enabled us to do. >> So, okay. So, there's a lot there. So, I want to get into the details of this. So, it's a really clever. So, so let's just let's just restate the the concept first. Okay, so what you are doing, you are measuring the uh outcome of your marketing spend on how much money it costs you to get a lead.

You have some target and I'm just going to completely literally make up a number right now. Okay, we'll call it $20 per lead. I literally have no idea if that's true. So, before somebody quotes me and says this is what Jackson pays per lead and you should pay it, too. Okay, just I just want to be very clear. I pulled that number out of thin air. So, uh let's just call it $20 per lead. Okay. Or $10. We'll say $10. We'll make the math easier. $10 per lead.

Okay. So, uh so you look at some target like that and then you decide uh whether or not your spend is effective based on how close you are to that cost per lead. >> Yeah, definitely. It's pretty much spot on. So, we pretty much track the value of each lead. We do a lifetime value curve where I know that maybe in the first 30 days a lead is worth in this example let's say $15 and then in one year it's worth $30 $40. So I understand the value of each lead and because of that I can set a target for how much I'm willing to spend for each lead.

So it just lets you the problem for our business using like a CAC is that it takes really long to decide to buy our product. We're a high AOB business. We're heavily seasonal. um you're either in the market or you're not in the market. So, it's pretty difficult for us to attribute spend to when it has impact. And we're we've just been stuck on this problem of like how can I better attribute spend to when it actually has impact.

The number one way we found is that it balances you sign up pretty quickly after you hear about our brand, especially if you're interested. And combined with that, we can track the impact in terms of revenue from that lead just exactly one to one. We just take our EMS um attentive, we match it up against Shopify, we see the email when it signed up, we see how much sales we got from it over time and we're just able to plot it over time.

It's been an incredibly powerful tool. >> Uh in what ways has it been powerful? So like what I really want is like how has it materially changed your behaviors? What what insight did that unlock for you that allowed you to to use it to actually drive either more or more efficient spend or both? >> Yeah. Um, I think it's always take a step back and I think taking off our marketer hat and putting on our consumer our day-to-day hat.

We got BFCM coming up. We know and I can guarantee for most brands, Black Friday is when you're going to see the most sales. The question I would ask is, is Black Friday when you think most people actually decide to purchase your brand? And we were talking about it over lunch, but we think that everyone has a little shopping list for Black Friday already in their notes app. they already have a good idea of when they're g what brand they're going to buy.

They just haven't bought it yet. So, what that thesis means is that during Black Friday, CPMs are going to be at all high, but you're not actually going to be capturing a lot of in or you're not going to be generating a lot of interest because you already have the interest. In contrast, I would say let's say early November, late October, this is probably the time where most people are actually deciding what they're going to buy. you're starting to get on their like we said their notes set.

The hard part is I think all marketers kind of understand this right now and we will talk about it but how are you actually able to identify that or quantify it or understand how much am I willing to spend in late October early November in order to reap the benefits in late November. And just the way we see it is we thought that leaves so being able to collect these emails is the most reliable way. The key part is you also have to understand the value of the email and you have to understand that even though what we found is the value of email is fairly predictable the timing of the value is very different where for example if I am advertising the week before Black Friday that email is probably going to have a huge spike in the next seven days.

If I'm advertising right now it's going to take a probably a month to get the full value of that email. And it just being able to understand that nuance and being willing to spend essentially when very few people are really spending and then being willing to pull back when everyone starts spending you start getting these inflated CPMs. It just allows you to get a little bit of an edge in terms of better understanding the actual interest of your demand. >> Okay, I still have a million more questions.

That's a great answer. Um so so again practically then did that materially increase your spend leading up to Black Friday like like because cuz what you know what I've I've told people before I actually have written and published some stuff about this uh in and some sort of external publications but but you know I had noticed a long time ago but back when I was a common thread collective noticed a long time ago that um that you know what we call the delayed attribution multiplier the idea of like the the way that uh value gets created from day one to day 28 on meta as the case may be, right?

The gap between those two numbers goes up a bunch in November relative to other times of year. And it's it's basically just the same thing. All you're saying is that like if you just measure on one day click on your meta spend and really on any marketing channel spend on November 10th, okay? Then uh then that number is going to get multiplied by a larger number uh during November than it would be like on a 28 day timeline in August, right?

So um so basically the idea is that on November 10th somebody again writes their note in their notes app or whatever and they maybe don't buy today but they do buy on Black Friday and that specific seasonal moment uh is unique to Black Friday. Um and so what that means is you should spend more on November 10th than what your metrics look like in platform. That's basically what it means is that your your spend should increase because the value of that spend is going to be higher than it looks like on day one.

That's that's the brass tax. So, is is that sort of the main way that this has been helpful to you guys has been as a way of sort of more precisely because what I just said is a little bit vibes based. You know, there's some you could do some math on it, but it's not super precise. Uh but with with an actual cost per email, you've actually got a way to sort of pin a a a leading metric here, you know, uh a leading indicator is what is what I mean.

Uh you got a way to pin a leading indicator to that spend. So is that what it actually got you guys to do is spend more at particular moments and spend less at other moments or spend more overall or like how did it actually play out for you? >> Yeah, I think a great example real example is tariffs in Q2. I think we all dealt with it. We all saw how volatile the market was and what we saw is that even though the buying intent wasn't there, there was probably a lot of uncertainty in the demand for the American consumer.

We saw that the intent in ter we saw the cost drop so low in our marketing platforms that we identified hey even though we're not seeing the sales to justify this spend level today we believe that over time these leads will mature to be valuable and we're getting them for the cheapest price we'll get them all year. So, we were able to surge into that and really build up for like say let's say our Memorial Day and Father's Day are our key moments and we just went in there with such a large fun.

We love talking about a funnel and we went in there with such a large funnel that when you have that kind of resilient baseline of demand that you just haven't realized, it makes execution so much easier. It allows you to take these big swings and not be worried about the downside. And I think that just keeps multiplying. So I think that's one concrete example is tariff volatility. Just being able to identify costs are low even though people aren't necessarily converting we're able to have a like you said a more precise way of measuring that demand. >> Well let me ask this.

So do you guys find that an email captured at uh at different times of year relative to seasonal peaks differently valuable? Because actually part of the thing that's sort of inherent in the thing that you're saying is like there's a lot of talk about in the sort of e-commerce advertising world about uh about sort of like how to create create how to generate creative that wins you know sort of all the time you know creating better funnels better offers better whatever it is.

Um, but what is sort of what people are missing in that conversation a lot of times is something that you're actually saying is central to the to the consumer's relationship to your business, which is moments, seasonal moments, which is a relationship between both promotion probably as well as, and I'm not sure exactly your own promotion strategy, but promotion as well as just like the reality of the surge in demand that happens at these different moments like you mentioned, Father's Day, BFCM, Memorial Day, etc.

Um, so, so what you're doing, as far as I can tell, and again, correct me if I'm wrong, is you're centering moments in your marketing strategy. You're saying like this is like we are going to behave as if the value capture is going to be disproportionate at those moments and our job is to make those moments as big as possible and therefore uh because that's where all the demand is. So that's where we have to squeeze the sponge or you know, like I said, you empty the funnel or what however you want to say it.

Yeah. So, first of all, did I explain that right? because I think a lot of people actually undervalue moments particularly in a business I mean in a business like yours they hopefully value it properly but if you're like a supplement person you're probably thinking less about moments but secondly do you find that the value of an email changes relative to uh the timeline to the moment so if I let's just take black Friday if I if you capture the email on September one does it have a different value than November one >> at black Friday >> yeah um so speaking about mo two parts of the question on the swan and how we execute against that We definitely are very seasonal because we're very gifting heavy.

We're a great gift for women to buy to men and we also have a large selection for men to buy to women. So that is definitely a core part of our business and it's always the same question of even though so a great example is our Christmas cut off. Let's say it's going to be late December, December 28th, December 21st. Um is that the moment where the spend has the most impact or is that just the moment where I love the phrase squeezing the sponge?

I think that's a great way to say it is, are we just squeezing the sponge of pre-built demand? And I think we both seem to agree that it seems like we're probably squeezing the sponge. There's nothing special about, you know, December 20th, December 21st. And then it becomes a question of knowing I'm not creating demand that day and I'm just capturing when can I actually spend. And that's constantly the issue we have.

Um, a great advice I have for a lot of brands, and actually probably borrowed it from a different brand, is just ask customers when did they first hear about you. See how it tracks over time. And if you're a larger ALB brand like us, you might see a shockingly small percentage of customers hearing about you within, you know, the past month, past week, and a shockingly large heard about you for years. Um, I love the example of like I always try to understand my own shopping behavior because I think a lot of the best ideas I've had have been from that.

Um, Hexclad, super well-known e-commerce brand, the he the pants cookwware. Purchase experience for that was I heard about it for probably two years ago. Then I saw it all over social media. Could not get away from it. It seemed like every influencer was using it and then I happened to get a gift card from Amazon and that's the moment where I decided to buy. And then I know that thinking about, you know, I came from this analytics background.

They have an analyst somewhere who's looking at me and just being like, "Oh, our spend probably had more impact today because Michael decided to purchase." But in reality, it's a complex story underlying the actual data. And it get it gets averaged out. But I think that you can't lose sight of that story and you have to be able to action against that story knowing that hey I might be spending a dollar today but I might be realizing the benefit over the next three months over the next year maybe even two years depending on you know your brand shopping behavior.

So that was my whole spiel about moments your biggest moment isn't necessarily your most impactful spend time and then it just turns into when is spend impactful. Um, the other point you brought up was about oh the value per lead. I think that's the tricky part to execute and we talked about it how this was something you would only recommend to brands that have it very dialed in. So what cost per lead does is it lets you better attribute spend to impact but it makes it harder to attribute impact to dollar.

So I love viewing as a spectrum where you have CAC on this side. I know exactly how much dollar my impact has. It's my AO ob, but it's hard to attribute spend. On this side, I have CPM where the impact is impression. I know exactly how many impressions I got and how much it to help to get costed to get them. But it's hard to attribute impact. If you talk to me about like how much is each impression actually worth in dollars, almost clueless.

So you have this spectrum and then from this side to this side you essentially have what you have a CAC you have add to cardart so cost per ad to cart cost per email cost per click and then um maybe some other like cost per view or cost per hook. So what we found is this cost per email isn't perfect but it lets you better attribute the marketing and it adds a complexity to attributed that impact which is kind of what we discussed.

So what we you will see is that there will be seasonality to your impact. Um for us a great example is immediately before big moments is when our value just skyrockets. This is probably when everyone is actually deciding what could purchase. On our actual big moments the value can be very high but you don't see that nice lifetime value curve. You essentially realize that value on the same day or you don't realize it for a very long time.

I think intuitively it makes sense. If I'm shopping for a Christmas gift and then Christmas passes, I'm probably not shopping for that Christmas gift for another year. So, it's really about just trying to it's really about trying to decide where you want the complexity. Do you want the complexity to be how impactful is my marketing spend or do you want the complexity to be how much dollar do I make for a certain metric?

We tried things like cost per ad to car, cost per click, these different ways. even cost per impression. Um, kind of glanced at that one and decided not to make it a guiding metric just because it's pretty nasy to do dilute it. Uh, yeah. And just the value of this email capture is that you're able to track that email in Shopify. So, I can see exactly how much dollars I got on that email and exactly when I got it, where I click with all this privacy stuff.

I think there are a lot of improvements with ability to have better resolution. Don't think it's amazing yet though. So, it's really what we landed on as like the best way to balance um that attribution of cost and dollars. Yeah. If you are liking this conversation, you can thank my friends at Intelligjs for introducing me to Michael. Uh this is a obvious fit because intelligence is all about empowering [music] smart datadriven uh operationally excellent e-commerce thinkers, operators to make datadriven decisions about the things that really move the needle in their businesses.

And they do that by allowing you to test uh well beyond the basics. So much much bigger tests than just simple split tests of uh headlines uh and button colors and all that kind of stuff. Those things can be useful. Certainly headline testing as a path to messaging testing can be really really valuable. But there's so much more you can do and you can do it with Intelligjams very quickly and easily. Intelligjs is very easy to install.

You don't even need a developer. And right away you can get testing things like offer price. You can get uh or your product price. You can get think test things like uh introductory offers, you know, the 10% off you offer to customers, etc. [music] Your free shipping threshold, your free shipping charge. You can actually run full like Shopify theme split test. So if you're launching a new website, you can test it before it goes live. intelligence to see if you're about to break your conversion rate by having a new website.

Um, just anything you can think of, they allow you to do that. They've actually even re recently introduced the ability to edit and test [music] checkout, which is amazing because that's very close to purchase and a huge value ad as well. If you're serious about testing the things that really move the needle in your business, [music] um, responding in a data driven way in your business, use a tool that literally every one of my clients uses, that is intelligence by going to intelligence.io.

Use the code Ferris 20. F A [music] R I S20 to get 20% off your first three months and get test get testing and optimizing [music] today. I'm curious if there's any insights specifically that you can share that this tracking yielded for you that were surprising like at some point did you did you guys by doing this discover something that was like whoa that's different than we thought. I think it's always about these pockets where there are weird I I think it's less so a broad insight, but it's more so where we have these random moments and I'm guessing it's a macro factor of everyone else pulling back spend where it's like why is marketing cost so cheap and sure our spend might get uncomfortable as a percentage of our total sales but we just feel confident that we're generating so much future value that those moments I would say even though we surge into and even Even though we have a lot of this data in back, we still look back every time and say, "Oh, I just wish we could put in another 10%, another 20%." Um, I think that like in terms of like how to actually execute this since it sounds like we're getting very tactical, the number one thing I can say is you have to understand your dilution.

And we talk about it a lot. Um, I love the I love the quote that statisticians sleep with one head in the oven and one head in with in their feet in the freezer so that the average temperature is comfortable and it's kind of the same where you know you can look at for example your value per lead but it's about what is that incremental lead and I think where you can go wrong is if you assume your incremental leads value is equal to your total.

Um, just speaking to how we look at it, for example, I can graph the number of leads on a scatter plot uh by day and I can graph how much dollar each one generated and I can see whether there's any kind of correlation between the number of leads and the average amount of dollar generated per lead and then I can kind of the great part is you can just visually see these things where I'm able to visually see at what point there looks like a break point.

Um, for us it's a lot of times like when we're starting to get, you know, hundreds of millions of impressions per week, suddenly it's like, oh, the US only has what, 350 million people. Probably if I'm getting 700 million impressions, we're getting a little bit we're getting a little bit into the repetitive diminishing returns feel. >> Yeah. >> So, yeah, >> I think that what's interesting is it looks fairly clear when you see these break points of like, oh, I hit this number of leads per day.

I hit this cost per lead. I hit this amount of spend per day and you just start seeing like, oh wow, that dropped off very quickly. Even though I'm getting, you know, leads, it's obviously the quality is being diluted. Um, yeah, I think the other one that has been really important >> and that's uncomfortable and I think it might turn into a whole different discussion about like how you make tactical decisions, but you have to be willing to make predictions as a decision maker and you have to be willing to look back and say, was that correct or was that wrong?

And that's something we push in the organization a lot is like you're expected to be wrong 20 to 30% of the time. have the permission and the expectation to be wrong, but you have to understand when you're wrong, why you're wrong, and just so you can learn from it. Um, I'm about to go on a super random tangent, but one of my favorite stories. Have you heard of Japanese chicken genderers? >> No. >> So, what happens is it's really important, and you'll see how like acts.

It's really important to be able to tell whether a baby chicken is male or female. The problem is it's for some reason next to impossible. So in Japan they have essentially these ch trained chicken genderers that just by trial and error like a feedback loop can do it accurate about 98% of the time. The really crazy part though is that even in 2024 with all of the AI, with all of the crazy computer vision, computers can only do it 81% of the time, 82%, sometimes maybe they can do 90% I'm hearing, but still these people just by pure trial and error are able to do it more accurately.

And the thing is they aren't able to actually describe why like they're putting in one pile or why the other. it just because they just pattern recognition in like the back of their head they're able to do these things and I kind of view connected to I >> like I think a lot of it is just like building that kind of feedback loop for yourself as a performance marketer as a decision maker >> so that you can start trusting your instinct just because you have you refined it in a pretty quantitative rigorous rate way over time I think that's really important so like when it comes to these cost per leads.

We say, "Hey, this is the cost per lead we're targeting. This is the, you know, impact we expect." And we just check it on a weekly basis. >> Yeah. And so you're saying the prediction the value of the prediction exercise for you guys uh is a way to test and learn and get feedback about sort of forces you into a pattern of getting that feedback both negative and positive so that you can then kind of figure out how to think better about the total mark about you know marketing mix among other things in the business. >> I think what it's taught us is we understand how our model breaks.

So we've learned for example that um the model is always a simplification of the real world kind of mentioned it with like my purchase journey with hex class but >> yes >> um >> the assumption that your spend today will generate a lead today >> and if it doesn't it does nothing and will have no impact on tomorrow is not correct. So I think that's the trap we falling in is that you just start relying so much on a model, so much on one way of thinking that you lose sight of the flaws.

So what this feedback loop has forced us to do is recognize more of the complexity that like hey my cost per lead today, how much it cost me to get someone to sign up per email is still dependent on yesterday. And you know what the crazy part is? is that so we always talk about like incremental rorowaz as a business like how much impact will my dollar today have >> it's like the idea of this is getting pretty out there but a non-marovian thing so loom's chat GPT they built on marovian chains which are that like you can show me some historical data and I can tell you exactly what the next line of text will be um if you think about it spend is not like that because my rorowaz today on a 30-day basis probably depends a lot on how much I spend tomorrow or next week.

So then in order to answer how much ROI am I going to get out of Zelar spent today, you also have to tell me how much are am I going to spend tomorrow and how much did I spend yesterday and just turns into this really really complex dynamic. Now I think of course the only way we as humans can navigate these dynamics are like these simplifications of a model like a CAC a cost per lead basis. So we choose cost per lead because we think that that's a more accurate simplification but it still breaks.

[laughter] gets a truth is that it breaks and you have to figure out the patterns of what causes it to break. So that next time you're in those shoes, you're able to understand like, hey, even though cost per lead is really low right now, we understand that maybe with the promotion ending pretty soon, we're not going to realize a lot of value in two weeks. Something of that level of complexity. So pretty out there. >> Yeah. >> No, no, it's good.

Um I I don't think it's out there at all. think I think actually it highlights something that we talked about or that you mentioned a little bit ago which is that like um this and this I think why you are an interesting guest on this show is that uh there's this assessment I think almost every operator needs to make which is like how good am I really and therefore what and and really not how good but like what am I great at and therefore what ought I build my operational processes around versus Not and I I actually think this is a sneaky difficult part and sneaky important part of a lot of brands like there are some brands that should disregard everything you are saying right now in this um in this episode because they do not have the analytical chops to do it. um and they and they they don't have the data gathering and and it will be a huge organizational lift.

And instead what they ought to do is redirect that time towards actioning as many possible swings of the bat as they can, you know, or whatever because they're really good at that. they're really good at building or or they ought to um reduce their growth goals because they just are like just need to get realistic about the fact that like the the more the faster and harder you push it, the more technical you have to get and the harder cash flow gets and and the more you're making the game difficult and you know or whatever.

Um or they need to contract somebody who can help them sort through some data questions because they recognize this is what it would look like and they need to add a team member or whatever it is. But I think the way you're thinking about this reflects this combination of um intelligence about uh tactics and strategy and about model building and analytics with what I hear is some operationalization of those tact of those strategies.

Right? So the idea of like you expect people to make predictions and then to be wrong at certain amounts of time means that internally in your organization there is some there's some operational process where somebody knows they have to do this thing where they make a prediction they run a test they do whatever and that means your ideas about prediction and feedback loops become operational realities etc. And and that that I think is the sort of very very difficult thing about a lot of these things is that like it's hard to figure out um how and where to put your to put your focus for your brand.

But what I'll say is in the midst of all that for a brand like yours that has this very core problem of uh the awareness and demand are often disconnected to each other, right? Because of different moments. Uh like this is a really smart way to go about it. And I think a lot of brands listening to or watching this probably ought to dig into this problem a little bit because those moments come and go. I mean, I I certainly have at least one extremely seasonal brand that is a client of mine and I've wondered for a while like how should we handle this?

And my thought is like, okay, what is this organization's capacity to do something like this model as a way of solving this problem? because I really do wonder every year if we are underspending in October. Um you know uh because our AM is too high in October and our spend is low and I know the CPM is going to go up in three weeks and a lot of those people are could actually buy and maybe we really you know so it represents a real tangible way forward but it's also hard.

Uh it requires it requires some some efforts and so sort of figuring out how to solve the organizational level isn't just a matter of papering the math. It's a matter of figuring out, okay, how do I build this into my organization and then find ways to sort of reinforce it and make sure it gets operationalized because that's that's it's really really hard. You know, it is time to get better customer service software that also costs you less and that is by upgrading your customer service help desk to rich [music] panel.

The key thing with rich panel is that it is built from the ground up with AI, powered by AI. And if there's a place in your business where it's just very obvious that AI can make a big difference really fast, it is the customer service help desk where AI can help your agents do a lot more faster and therefore create a better experience for your [music] customers and critically uh your agents uh or or a rich panel at the same time has a customer service >> [music] >> uh self-help portal powered by AI to help customers uh get answers to their questions really really fast.

And you put that all together and what happens is that on average when people install [music] rich panel, especially coming over from Gorgeous or Zenesk, they see a 30% ticket reduction, ticket volume reduction [music] uh in tickets because customers are getting answers to their questions faster through the self-help portal without having to go through long forums [music] or, you know, um Q&A sections or FAQs or whatever it is.

And um uh help centers, that kind of thing. And at the same time, and here's the amazing thing, Rich Panel guarantees that they will save you at least 30% on your bill from switching to Gorgeous [music] or Zenesk. So, you're going to have a better experience for your customer. It's going to cost you less money, and it's even easy to install. They know you don't have time to waste installing and uh training your team on how to use Rich Panel.

So, they promise to get it done with you in two weeks so that you can be up and running and going. Really awesome. Multiple of my clients are using Rich Panel. Multiple very, very large brands are using Rich Panel. It's not just cheap and for small teams or something like that. It is affordable, but it [music] can be scaled with very large teams. It even has an AI powered agent that can go and interact on your social comments, on your ads on [music] behalf of your brand. does an incredible job doing that.

Doesn't just hallucinate all the time or anything. You It'll do a really great job. It's one of those things that's extremely annoying about running Facebook and Instagram ads. You know, you should be interacting in the comments. It's really hard to do, really hard to stay on top of. Let their AI agent do it for you. Uh go to richpanel.com, richpanel.com, get set up today. You can follow the link in the show notes for that.

Panel.com to try it out today. Honestly, I love the direction we're going down because if we start getting into like how do you build great decision making into your decision makers or into your organization, how do you embed that? And I think that where I started from and just kind of what I've learned from my own experience is every bad decision I've made or that I've seen an organization make just really two issues.

It's number one greed or number two fear. And somewhere you're on that spectrum, the hard part is understanding and having the self-awareness to say, am I being greedy right now or am I being fearful? Am I leaving things on the table because I'm afraid to roll the dice or am I starting to get a little bit delusional about what the potential is and because of that I'm trying to chase everything when I just need to chase something.

And I think that's really what it comes down to is all of these like operational like be wrong 20% of the time, track the percentage of times you're right, track how often you're right in the correct direction beyond magnitude, it just comes down to building that level of self-awareness. Because the thing is that as soon as you have that ability to understand, am I being fearful or greedy? If I know I'm being a little bit afraid, I'll force myself to be a little more uncomfortably aggressive.

But if I see myself getting slightly over excited, it just turns into having the emotional control to be like, "Okay, we need to reel this back because I recognize when I was in this position two months ago." Um, I bet it all and maybe I lost it all. So, I think the other one we you brought up was just how a brand should look at like their spend on a macro level, on a year-over-year level. I think the number one and I can tell you about how we kind of viewed it in order to kind of justify some more risky more aggressive moves. >> It's about asking yourself the question.

I love thought experiments and the thought experiment is let's say hypothetically this brand it's been able to you know scale up pretty well 20% 30% 40% somewhere in that magnitude of year-over-year increases in spend asking why was I able to spend more this year and see the financial impact that I did versus last year and the second one is next year do I think I'm going to spend more or less than I am this year I spent 30% more this year and I'll probably spend 30% more next year and then it just comes down to like why can't I do 30% more this year instead of just waiting for next year and there are good answers and the three answers that we've seen internally number one is it's macro where you're able to spend more because um your Stanley cup and a car burned with your cup inside of it and you can grow to a billion dollars.

So that's number one, something out of your control. Number two is you managed to improve efficiency somehow whether it's you reach a new audience in an incremental way through a new category or you just got better at creative you got better at media mix number three which I think 8020 is the impact is just time for a brand like us where we tried to we still just naturally too closely tie spend to sales when in reality it's not just one month it's not just three months it's probably years and just we're forced because of like just the corporate culture in America, you know, drive for results that like we can't think on a 10-year time frame.

It is like an annual probably at most for most brands and knowing it's time it turns into a question of less so can I spend more or what's optimal but what's within my risk tolerance because the buyer you play with when you start you know scaling up spend quickly is that's a cone of uncertainty. Maybe my expected value, maybe my average is, you know, higher as I spend more, but the risk of a very catastrophic outcome is also higher.

The risk of a phenomenal outcome is also higher. And just asking yourself it it kind of shifted the question in our organization away from am I being optimal because the answer was no, we're not being optimal. We should probably spend more because we're going to spend more next year at worse CPMs. The question became how much can we spend to fit within our risk tolerance? And that's a very very different way to view the business.

It's turned into it just changed how we make decisions because then you're talking about then you can attach numbers of like hey um we don't we can't crash the business. We don't want to have a cash flow issue. We don't want to crunch the business. So that's outside of a risk tolerance. How much can we spend while you know setting ourselves up for this upside while minimizing that downside risk? It's been a pretty interesting thing and I think it's something that every brand needs to spend time on is if I've been able to scale up the business exactly just chart it out what has enabled you to and if you see a giant blank bucket of like hey we had a lot of AB tests um maybe drove 10% but we grew 40%.

That other 30% is probably just this time element that I think we all underestimate. >> Okay. I agree so much with what you are saying here Michael. There is a risk tolerance thing. I have been actually one of my big things recently is the idea that reasonable expectations is like a massive advantage in e-commerce. I've I put out some content about I've done some interviews about it recently, but there are just these people building really really really good businesses um that are um not that big yet relatively speaking.

Uh and it's because it takes a really long time for most people to build a really big business. there just there aren't actually very many people who can build a huge business very fast. The guy with the single biggest business in e-commerce that I have talked to lately is Sunil Agraal um who runs a hold co publicly traded in India called VGL group. It's uh his the it's a it's uh got $400 million business. There are six brands in it.

Um his biggest brand is 240 million a year. A monster business. He started it in 1980 and so my conversation with him is really really good uh because it um it sort of highlights like what does it take to do that over insane lengths of time. Well, what you end up with though is like a really really good business and without sort of ridiculous expectations about what to do next. And so um so yeah, your assessment of risk tolerance and then also sort of trying to factor in why are you actually growing the ways that you are and and what is actually causing your success uh versus not macroeconomic.

I think actually product development is really underrated in this just that like you release a product that that creates step changes in your value creation. um all kinds of things like that that I think are um are are harder to see when people sort of just want to find the meta tactic that takes them to the next level and figure out how they can unlock the funnel, you know. Um and I I just don't think that's quite right.

So, um yeah, >> actually listened to that one um just because you mentioned in some jewelry space and I was like I need to hear this and you my first thought and I think it's interesting for us as like mainly digital marketers. If I look at the top 10 brands right now like Nike, Lululemon, all of them got to being top 10 before digital marketing existed. >> And that sometimes it feels like we got gifted all this information.

We got CPM, CPC, we get daily creative analytics, we can have every metric we can think of. And it feels like that has actually made us like paralyzed by the amount of information. And that these OG brands who just view it as like marketing as spend on marketing and be ready to view it as like a risk benefit, view it as an investment versus a short-term like day trading vibe, which I feel like directionally a lot of DTC has gone to.

I feel like there's something there where there's not really a 10 billion company built on the back of Facebook, but all of our companies are built on the back of essentially Facebook. And it's like, well, all these companies were able to get to 1050 billion dollars without digital marketing and then how come we obsess so much about it? And I think there's good reason. It's simply the most powerful growth lever at this point.

Instagram advertising, it's phenomenal. The data allows us to iterate at a speed that no one else can iterate. But then in the back of our head, I think that sometimes we just need to remember that the biggest brands in the world were built on pretty basic budgeting, pretty basic marketing, pretty basic attribution. And that all of this information we're given, sometimes we need to ask ourselves, is it really enabling us to make better decisions all the time?

Is it allowing us to make better decisions in some ways and not others? just was it was really interesting because they just do TV, right? And if you ask me, could I make a what nine figure jewelry brand based on TV? Like there's no way. >> Yeah. Well, it's funny though because what Sunnil will tell you at the same time and what he's I asked him like, "What do you want me to promote in this episode?" You know, and he said he said, "I'm always looking for good digital marketers." So, I think on the one hand, you're right.

He recognizes that, you know, TV was a unique opportunity. I think they have their own channel or something you know um that like uh that it's an important opportunity. On the other hand, I think he also recognizes that that that tide is changing and that uh he will have to adjust that and that's why at the same time as he's running a $240 million brand. He literally started a brand new brand from scratch on Shopify like zero dollars on start, you know, on they're not just acquiring or whatever, you know.

Um and I think it's and and he said like send me digital marketing talent. That's the thing I want. Uh, so, um, so I think there is a real mixed thing here where it's like you I think you're right that there's something about I I think the thing that you're hitting that I think is right is that um that meta for people can suck them into the day trading mentality because it feels like day trading and it and probably just like actual day trading.

I bet there are people who can like make money doing it, you know, and make like a lot of money doing it, you know, but the thing that that everybody understands about day trading, I hope, is that that group of people is extremely small. Um and uh and that like the the best bet statistically is not that you should try to be on that extreme end of the bell curve of outcomes. Uh but that you ought to instead look towards the middle of the curve and say like well what is the actual best way to do this?

And it's not day trading. It's not a day trading mentality. Um and but meta I think yeah sucks people into that mentality a little bit and can become very quickly a get-rich get-rich quick scheme, you know. Um and and I you know this comes back to your comment about the the mistakes you've made have come back to greed or fear. Like this is the greed component of that. And I've watched this happen so many times. People just want that want that want that when actually like the core problems in their business are not going to be solved with one little tactic here and there.

It'll be solved with the actually like your cost per lead example is a great example of aligning the tactic to the customer in this really significant way which is like you understand what's happening for your consumer and therefore cost per lead as a as a as a way to track what's happening makes perfect sense because you know that they're going to be motivated to buy differently at different moments that their awareness and their demand are going to be in different spheres.

But you need some way to track that because the attribution is a mess and and on top of that there's privacy issues. And so you you sort of start with that lens of who the customer is, what they like, what they care about, and then you know how you do your your promotional strategy relative to those moments and how you do your tactics relative to those moments. And you think about doing that repeatedly over a very long period of time.

And that actually creates this. And I just love the way you're talking because you're just your your and I'll just tell you like your mental horsepower is just way higher than the vast majority of people that I have talked to in our space, including myself. Okay. So like just no question. Um and and you had your analytical chops as well. you know, you've got real background in these kinds of things. And yet your mentality is not like, "Okay, I'm going to go use that to go like build the craziest, fastest, biggest thing really fast." Like, uh, you should you should take that and apply it to reasonable expectations over a long timeline and then you'll get really really rich if you do that, by the way.

Um, but like uh but yeah, that's I mean, I don't know. I don't want to put words in your mouth, but that's the way I would sort of repackage part of what you just said. >> Yeah, I think a lot of it is just uh we talked about it, but I just love reading. Um, >> yeah, >> you look at my grades. I was a pretty decent student, but not breathtaking. But I think that a lot of it is just from reading the biographies of these really impressive business people.

And you know, like I think a great example just Warren Buffett, the man barely knows how to use a computer, yet somehow he's just destroying the stock market in terms of performance over what 50 years or so. And it comes down to he put together that if you get the tiny details wrong of like oh transaction value well like do I buy it today or tomorrow like you can get those tiniest details wrong but if you set yourself up for the big picture you can win pretty consistently and it's it just takes time and that the trap you can fall into is just trying to rush it.

It's super interesting just seeing how many different ways there are to succeed where we talked about it but like Ridge they crushed that they there seems to be some kind of process power there in terms of ability to execute on digital marketing and like these category launches like there's some kind of magic there and then you have these probably more like strategic great players like um we were talking is a great one probably where they make they've made it seems like consistently strong big picture decisions Whether it's like the head count, whether it's like their channel strategy, like product development, brand strategy, they nail those big pictures.

And then just trying to figure out for your brand where your skill set lies if you if you win and whatever winning looks like. You become number one in the category, you become a billion dollar company, whatever it looks like. What is the path you get there? Is it by these hyper tactical decisions or is it by just getting two or three major decisions right? Um, I think Ruggable is another great example where they just got the product right, they got the product right, they got the design right, they got the value proposition right, and then everything falls into line where they're great at that one thing and good at everything else.

It's impossible to be great at everything, but then like for your organization and maybe even on agency level, it's like I know you were talking to Taylor about like what are you great at? What am I great at? It's about within the brand having the self-awareness to be like what are we actually great at? What are we not so good at? Are we so bad at what we're not so good at that we need to uplevel it or do we push further on what we're great at?

So, >> yeah. >> Yeah, >> it's good. All right, we only got a few minutes left. Um, I actually want to ask a few more very very tactical questions about the cost per lead thing really fast. Um, just because I didn't want to miss them while we have the conversation. So, um, although I actually love that this conversation went from hyper tactical to, uh, to big picture thinking, I think the best thinkers are in the space have a really good ability to do both.

So, um, okay. So let's talk about a couple things in a technical sense in terms of your optimization of your ad spend. You are not optimizing for cost per lead, right? You're optimizing for conversions than tracking cost per lead. Am I correct about that? >> Yeah, that's correct. So cost per lead is a leading indicator for what we expect to be contribution margin is our end goal. So we tune our cost per lead goals to optimize for long-term contribution margin growth.

Let's say over 30, 60, 90 days. But you're running your ad spend as like standard. Let's just take your meta spend for example. You're you're optimizing for conversions, highest volume, highest value, manual bid, whatever you do, I don't know. But like uh something like that. And then just out the tracking of that spend, you're tracking the cost per lead. So I just because it's an important clarification. I'm not telling everybody to go optimize for leads on Meta.

You will get garbage leads if you do that. I'm pretty sure. >> No. Yeah. Um, [laughter] I've actually been bugging our team to maybe try it out once or twice, but I don't think that >> I would too if I was you. I would allocate some amount of spend. And my hypothesis would be that it would perform really poorly. I would I would put a giant um I would put a giant like haircut on what the uh the actual value of that lead is.

But like I would reduce it by like at least 90% honestly. But look, if you get the cost if you get the lead for um 8% of the cost of the other one and the value is 90% less, you win, you know. So, uh if you if you get the cost if you get the the and anything's worth paying for if it's cheap enough. So, um so yeah, even if it's a really bad bad quality lead. So, yeah. I Okay, but right now you're optimizing for conversions.

You're going to try optimizing for for CPL, but um but yeah. Okay, great. Does um your value per lead change relative to the offer with which you capture the lead? For example, if you're doing a standard 10% off, you know, for new customers versus 20% off for new customers or something like that, does the value per lead change? >> I love that question because I I think that I could go on another rabbit hole of CRO testing. >> Yeah.

LTV over CAC. view it that way where if I do 20% off my CAC is going to go down, my LTB is go down is going to go down because of the discount. Which one goes down faster matters whether or not it's a good idea. It's not worth it to guess. I think that every brand based on your unit economics have a different situation, but I think it is one of the most important things to test. >> Yeah. Yeah. And I well part of what I'm wondering is if you guys are actually assessing the value difference in a really clear way because you know the thing people everybody worries about um with almost any leading indicator is something you sort of noted earlier which is degradation to value based off of any number of factors you know um and so I'm curious sort of how you guys think about different tests like that and different cohort value creations based off of like offer price or whatever it is.

Yeah, I think that when it comes to offer price, it just about measure the dilution, see what the AOV, see what happens in different silos to the AOV to the sign up rate to the number of conversions and you'll see some kind of interplay with them and then run a statistical test on each one separately, the key part is separately, and then see which ones are noise versus truly happening. Um it's pretty surprising where for example I think some brands are going to find not the majority but some will find that as they increase their discounts their AOV might actually go up because suddenly people are unlocking a higher purchase >> right >> um on the other hand some might see the opposite where they increase the discount and conversion rate goes down because people there's some level of perceived value people think that it's over discounted so it is very personal to every brand but it is one of the most important levers to pull I would say definitely testing your offers, your signup offers both on the site and also through like email. >> Last question about this.

Are you optimizing are you doing anything to sort of or or just given your focus on cost per lead, are you doing a bunch of stuff to increase the value per lead at the tactical level? What I'm really thinking of here is like changing your, you know, split testing your flows and things like that so that once somebody actually signs up um with their email address, are you guys doing a lot of sort of posts signup testing uh of of tactics? >> We're actually hiring a retention person right now.

So that is the goal. I would say the other part that sometimes gets misconstrued is that cost per lead is the indicator where the goal isn't to get leads. the right thought process is that if you're getting leads, you have high quality traffic, right? >> So, um it's more of like a canary in the coal mine that's correlated to strong performance but isn't necessarily your performance driver. I think it goes into what you mentioned though like optimizing for cost per lead where suddenly if I rig the canary then it becomes useless.

It's only useful if you give it like if you view it as an indicator and not actually test the metric or the driver. Yeah, I'm glad you said that because that's that's clear to me from talking to you about it that you're not actually saying that it's the driver. You're just saying it's a really reliable indicator. What you started the conversation by saying is you said the reason we like this is because you can tie uh the lead to the exact moment of the lead capture with the email address and then to the exact moment of purchase with the email address.

And because those two things line up so perfectly, it just becomes a reliable data set. That's that's the reason it's so useful. Um, but I would kind of wonder if that if for that same reason, even though it's not the driver of value, for that same reason, it becomes useful uniquely in like post capture testing because because you're just testing something that you have such reliable uh data for. So that you know it's again it's not that like it's it's uh um it's not that it's particularly about the way the email address drives value but just again because you're tying the test to the email address and you already have all this measurement set up against that it might yield this is my theory anyway it might yield interesting test outcomes or something like that you know I don't that was the thought >> oh I definitely agree I I kind of I think that you can view the Jackson organization as just two distinct functions one is optimizing cost per lead so this is like our creative optimization, our performance marketing, our media buying.

They're trying to get cost per lead without just ragging their deck and doing something weird with like your campaign optimization. And the other side of the house, let's say, is like e-commerce, pricing, product, all these kind of things are about optimizing the value per lead. Like every or every part of your order solution can play into it where your CX team might be increasing your repurchase rate, which would increase your value per lead.

It just simplifies an otherwise 20 directions into two directions. Does this project increase cost or improve cost per lead, improve value per lead? Or something like a product launch is actually you brought it up interesting because it kind of does both where you have an LTV play, you have a cat play. um category could both >> reduce your cost per lead by reaching a new incremental audience and at the same time you're getting a higher value per lead whether through AOV or repurchase rate or just through uh yeah I guess AOV and repurchase rate would probably be the main two I'm going to ask you one last question and you can make it as brief or not as you want.

Is there anything that's just sort of top of mind hot on your mind right now in the e-commerce space that you think is just worth sharing? it could be totally unrelated to the conversation we've had so far, but just something you've been thinking about noodling on, noticing, something like that that's worth sharing with an audience of e-commerce operators and media buyers and those kinds of people. >> So, I think what's top of mind for me and we see a lot of brands breaking into different spaces obviously like gummy supplements, obviously like CPG, um, a lot of different routes to go and kind of talked about it.

What is the competitive dynamic within D2C as it gets more crowded within these categories? Because I think that you can view business categories from a macro lens point where there's a rapid expansion and then almost every time there's a contraction. There used to be 50 million different airlines and now there are what like five there used to be 50 million different phone companies. Now there are what two Apple and Android.

What does the contraction look like for DTOC whether and for categories where it's always top of mind for me where I'm in a new category and it feels like we went through this rapid expansion phase where a million companies got into men's jewelry and then it feels like right now we're hitting that contraction stage. I'm super interested in some of these big rising categories namely like supplements CPG. What does their contraction phase look like? what does a competitive dynamic look like as more people crowd into this space because it just looks like a money printing machine like we were both talking about it but I it's like well I need some a little bit of exposure to that side of the house uh subscription based businesses um I think it's interesting from like a macro level what will happen um and I think that like obviously these brands are looking for differentiation they're building moes around themselves and just how sturdy are those moes going to be against rising Facebook costs, more competition, maybe even more regulation.

Like, super interesting. >> I love it. Um, great thought. Uh, I wondered if you said something really tactical or if you said something bigger picture, you went bigger picture. I like it. >> You are trying to get more active on Twitter. You said on X. Um, so I have already I have already uh linked out to um to you on my feed, but Michael and Eggy, E GI, that's your dog. Uh, on X. Go give him a follow. I've also added you immediately to my Titans of DTC list on X.

So, if you're following that, you'll already get Michael's tweets. You're also on LinkedIn. Um, Michael, thanks so much for your time. Uh, all of the links for, uh, Michael's X and LinkedIn are in the show notes. I'm sure I will have Michael back at some point if he has time. Um, thanks, man. Appreciate it so much. >> Yeah, definitely. Super appreciated. Definitely need to catch up again pretty soon. >> Let's do it. We live close. >> I hope you see what I meant in the intro. that Michael is a great conversation partner.

Guy really knows what he's talking about and I love his ability so much to go back and forth between highly tactical and big picture thinking about business. Um, very curious mind. He told me when we were having lunch, he reads two to three hours a day, plays piano. A lot going on there. I just think it's the kind of mind that I notice stands out as somebody really, really smart. So, um, anyway, super enjoyable conversation for me, a guy who I definitely hope to have back and who I expect if I was doing predictions, uh, you will hear more and more from throughout the B2C space because he's just really smart.

Go follow him in the places that I told you. The links for that are in the show notes to go do those things and, uh, and I think you will, uh, get value out of his content as he continues to put time into that. Uh, and of course you can reach out to me at podcastfgrowth.com at afgrowth.com for everything that I'm doing [music] uh over there if you want to think about working with us in the future. Um, you can do that in the next year.

We will be opening up more space for clients. Um, so yep. And then uh what else? You should subscribe wherever you're watching or listening. Of course, don't forget to do that right now. If you made it this far in the video and you're not subscribed, what are you doing? Subscribe. You're going to like the show. You're going to like more of the episodes we do. It's going to be good. That's it. Um, oh, Rich Panel and Intell Gems.

Let's not forget them. Thanks so much to both of them. They're great sponsors. I really love working with both. Go check them out. Links for those are also in the show notes in the description. Uh that's it. I'll talk to you next time.