• Learn how to get an accurate picture of which ad channels are performing well.
  Who you need to exclude from your prospecting audience.
• Facebook, Google, TikTok, and YouTube… which platform is best for your business?
  How to “feed” Facebook for optimization.
• Discover how to scale an Ecommerce business through paid media and what you MUST do before you start.
  Why segmentation of your email list is CRITICAL for success.

William Harris: Usually Facebook’s performing significantly better than people realize it’s performing. So if they can get to the point where they understand attribution better, it is still the best one and the most mature and the most stable of all the ones that you’re looking at for that. Google Ads is, is by far the best for capturing existing demands.
So those two, if you’re only going to do two, those are still the ones that we 100% recommend.

Russell Miller: Welcome to the Heroes of Ecommerce Podcast, hosted by Russell Miller of Ryzeo. On this show, we speak to ecommerce business owners and leaders, the unsung heroes taking their industry by storm. Tune in as we share success stories and discuss ecommerce ideas to take your business to the next level.
So let’s dive in. So I am Russell Miller GM of Ryzeo, and this is Heroes of Ecommerce and with us today, we have William Harris who is head of the Elumynt Agency out of Minnesota. Very excited to have William with us today. He is a, his agency is a master of advertising, particularly for ecommerce.
And so we are going to provide you a lot of value this time. Learning about what you can do, you know, in advertising especially. As we record this, we’re in November and so Black Friday’s coming up, so that’s very exciting too. So yeah. So let’s start off with an easy one. William, maybe, maybe you can just, you know, give us a little bit of background about, you know, you know, who you are and what you guys do at Elumynt.

William Harris: Yeah, thanks Russell. I’m excited to be here. Yeah, I am the CEO of Elumynt and we are a rapid growth ecommerce agency. We’ve helped 13 of our clients get acquired. Many of those were for upwards of 50 million. The largest one sold for about 800 million, so we’re not messing around. We were just ranked as the 12th fastest growing agency in the world by Ad Week, and Google hand selected us as one of only a couple dozen agencies for the Mind Gap program, for their International Growth Agency program.
And I guess the biggest way that we accomplished most of that would just be through advertising. So Meta, Facebook, Google, TikTok, Reddit. We’ve developed a series of systems for optimizing ads around EBITDA profit, instead of ROAZ or other metrics that we just don’t think matter.

Russell Miller: Now that’s a really interesting one because, you know, I think, we’ll, we’ll get into this a little bit, but, but it sort of caught my eye as I was kind of doing the show prep that, that you’re focused on, on profit, which is something I think most ecommerce stores really kind of like viscerally get instead of ROAZ which, and, and maybe you should actually explain what that is, cause all of our, our listeners may not know.

William Harris: Sure. So EBITDA is very much like a tax accounting thing for earnings before interest you know, deductions taxes, deductions, amoritization. And then ROAZ is return on ad spend.
And the way that those have been typically defined. If you think about ROAZ return on ad spend, it is a beautiful sounding metric. It’s literally the return. It’s like ROI, but it’s the return on your ad spend. Yeah. And I think the, the biggest issue that we ran into with this is that it, it has always been a misnomer.
It is not, it’s never been appropriately named. It is the re it is the reported return on your ad spend. Now, if you go back to before the iOS 14 issues, oftentimes platforms would take credit for more than what they incrementally technically drove, right? So a lot of people were just like, oh, I don’t believe what Facebook’s telling me, because I think that it’s taking way too much credit for stuff than what it’s really driving.
So the, it was a ROAZ but it wasn’t a ROAZ that you could believe. It wasn’t the actual return on your ads spend, it’s just a reported. Well, then iOS 14 crushed it the other way to where it’s like not reporting as much revenue as maybe what it could take credit for. And so that’s still, it’s still not actually the return on ad spend.
And so I think calling it return on ad spend is just it, it confuses people into maybe thinking that it is something that it isn’t. Whereas profit EBITDA, this is a fact that you can arrive to, with your CPA, your CPA’s gonna know for sure whether or not you actually hit this number, and I think when we think about what we are doing as a business is customers don’t pay us out of their revenue.They don’t pay us outta their ROAZ, they pay us out of their profit, out of their EBITDA. And if that’s not there, then we, you know, did something wrong. And there’s a lot of times where I have seen you could increase your ROAZ with, without increasing your profit, you, you could actually lower your profit.
And it seems maybe counterintuitive, but without going too deep into it, at this point in time, let’s just say it’s very easy to increase one without actually increasing the other.

Russell Miller: And so by focusing on profit, you’re, you’re helping them focus on the bottom line thing that you know, really matters to them.

William Harris: This is ultimately what matters to the company. And, and sometimes one of the things that we’ve found is there’s maybe a lack of communication oftentimes between the finance teams and marketing teams. And so for lack of a better idea, let’s call it for marketing instead of marketing, right? Finance marketing.
But, but a lot of times the, the, the marketing team even themselves are not even aware of maybe what the EBITDA was or profit was. Cause maybe they’re an employee and so the P&L isn’t being shared out to every single person in the company. And so they’re operating around their own set of assumptions and guidelines and rules too.
And what happens is, and this is we’ve run into this, there was a, a customer, we increased their, their EBITDA by massive amount, and I’ll get to the number here in a second. Right? First month that we took them on and they, basically, we ended up charging, I, I wanna say it was, you know, a little bit more than what they were used to double, what they were used to paying the previous agency, let’s just say for, for number’s sake, it was like they were used to paying 10,000 a month and they paid us 20,000 that month.
And their marketing person, you know, is only aware of the cost. He’s only seeing this and saying, wait a minute, wow, that’s a lot more than I’m, we’re used to paying, like, I’m gonna get in trouble for this, right? And I said, well, wait. Let’s look at what happened. When we subtract your cogs, when we subtract your overhead, we subtract your ads spend, we subtract the agency fee, we check, you guys made $800,000 more profit in a single month.
Was it worth paying $10,000 to the agency to to, to go ahead and, you know, do that? And they’re like, well, yeah, when you put it that way. Yeah, I guess it is. And so by getting the finance teams and the marketing teams aligned, we found that it allows everybody to be a little bit more on the same targets.

Russell Miller: That is huge.
And you’re probably doing sort of a bring a, a big service for the management by kind of bringing those two silos together to sort of see a full picture. Yeah. Cause often those are just, just totally separate and they never really, you know, they never really talk.

William Harris: Almost don’t wanna talk. Yeah, right.

Russell Miller: That’s probably true. So we are, we are going into high season right now for most consumer focused eCommerce businesses. I know that a lot of our customers out there either work with smaller agencies or, you know, you know, have someone internally and try managing themselves. There’s probably like a real spectrum.
What are some common ad mistakes you see ecommerce companies making?

William Harris: Yeah. Okay. So. Right now, and this is especially important because of, I think everything that we see going on with the recession that we’re not quite allowed to call it a recession yet, whatever we would call this, but like the, the impending ness that we have where, from an advertising perspective too, making sure that we’re aware, you know, Facebook’s down a little bit, Google’s down, Microsoft’s down.
Like all the digital platforms for the most part are seeing some decline there. Shopify, right? Laid people off. You mean in traffic or decline? How? In, in, in ad spend especially, which is where I, I’d say like if we’re talking about you know, looking at from even their stock price, you know, investors and everything too.
So, which is why a lot of them, they’re, they’re laying people off and things like that. So there’s a, this is a very real thing that we have to be aware of. So you have to spend each dollar as intelligently as you can. And I think one of the, there’s two, two common ad mistakes that I see right now that I think are absolutely destroying some, some companies.
The first one is gonna be a poor understanding of which ad channels are actually performing. . And I, what I had been is I see there’s too much focus on the creative, which is important. And so I don’t wanna say that it’s not, it’s still very important and can make a big, big difference, but I think that there’s maybe not enough intellectual discussion about the actual analytics and the algorithms that are going on here.
And it’s not to say that people aren’t talking about attribution. They are. But I, I feel like they’re talking a little bit out of their, their butts. Like ASCE Ventura, I don’t know if you remember the ASCE trip part where he kinda like bends over, he’s like, excuse me, I’d like to ask you a question.
That’s kinda what I feel like’s happening here. Because attribution was a problem before iOS 14. It was after iOS 14 as well, used to take too credit.

Russell Miller: And so, because you mentioned several times and, and I’m aware of it, but I, I think it would be good to just, you know, do a little sidebar.
What changed with iOS 14 that, that people should know about.

William Harris: Sure. The, the quick change with this and, and everything gets lumped into iOS 14. It actually started before that ITP 2.3 came out like six months before iOS 14, and that created a big issue as well from a tracking perspective. The, the biggest issue that I think people see from iOS 14, there’s two.
One is the attribution side of things. So, you used to be able to get immediate feedback on what worked so they, Facebook, would know whether that purchase came through and so they could take credit for that. Apple basically delayed that as well as limits the amount of information that’s being passed through so they can’t tie as much back.
So the way that I’ve described it to some people is imagine if you had 10 kids and you told all 10 kids to go clean up their room. And normally five of them come back and say, Hey, I cleaned up my room. And now you, and you go and you inspect, and sure enough, you know, five of ’em clean up the room. But now you say, Hey, go clean up your room, and only one comes back and tells you that they cleaned up your room, but you go and inspect and five rooms are still clean. You’re just getting less reported back to you. It doesn’t mean that anything has changed necessarily. You’re just not getting the feedback into account for it. So that was the primary issue. The secondary issue is it does change some of what needs to happen within the platform, and this is where I think a lot of people have missed it.
Is they’re not making the changes that needed to take place. And this is where we’re getting to the two parts then. So the, the, the one part here is fixing the attribution. One of the ways that we’re really talking to people about doing this is get rid of this idea of a single source of truth. It’s, it’s, you, you, it has never existed and it will certainly never exist now. It’s always gonna be flawed in some way. And so you need to look at it more three dimensionally and understand what is the platform saying, but also what is Shopify saying? But also what is Google analytics saying? But also how about using post-purchase customer surveys.
And so one of the articles we just published last week, we found, we use a, an app called Fairing. F A I R I N G.co and we found through using that, that Facebook was taking credit for so much less revenue. It was under reporting by about 80%. It was massive. And that obviously unlocked significant growth for this company that thought that they were not doing very well, but actually were doing well. I just needed to scale up. So that’s the biggest ad mistake that we see is just not understanding which platform is, is working well. The other one then is as far as not setting things up correctly. I’m, I’m sad to see that I still see this, but it’s probably the biggest thing that I see for, for smaller companies and even a lot of people that are working with larger agencies, is proper exclusions and inclusions from an audience perspective within your ads.
You can imagine, let’s say before, well, with iOS 14, they can track the data on those purchases then for up to about seven days for most of ’em, but, but only for about one day. So after that, if you are not excluding people properly enough, if you’re simply saying, Hey, exclude people from my prospecting audience based on 180 days on the pixel or something like that.
Which is what we used to do. After somebody’s maybe purchased two days ago or eight days ago, they’re all right back in your prospecting audience. And now your prospecting audience isn’t actually going after net new people. It’s just retargeting. And so you might be spending more money on that thinking that you’re going after net new people and you’re not.
And that’s what we found. Almost the biggest change is literally making sure that you have the proper exclusions in there, which I’ll get to in a second. All of a sudden your prospecting is really prospecting and, and it’s almost night and day difference in terms of what happens to, to the platform. And one of the best ways to do this is to, to make sure that you’re including from an exclusion standpoint email addresses.
You can do this dynamically through Klayvio, as well as other programs. But also go ahead and exclude anybody that has interacted with your Facebook page or interacted with your Instagram. So now you’re truly going after just net new people and you’re trying to exclude as many people that have previously engaged with your brand as possible.

Russell Miller: And so just so this makes a lot of sense to me and I’m, I’m gonna try to restate it maybe in a more simple way for our customers. You know, your, your prospecting is very different from your retargeting. Your prospecting is, you know, completely brand new people. They’ve never heard of your brand, and if you’re not careful, what you think is prospecting spend going out for new people can actually be going out to people who maybe have just purchased from you if you’re not careful and excluding them from your, your prospecting campaigns.

William Harris: Exactly. And the reason why that makes such a big difference, because I. Let’s go to the, the attribution side of things. A lot of people will see in their, their campaigns especially if they’re a little less savvy on what’s happening here. And we talked about like high ROAZ doesn’t always mean incremental revenue. Retargeting has a higher return on ad spend in platform for obvious reasons because they’ve already interacted with you, right?
But here’s where the problem with that comes is if you think that you’re spending a higher amount on prospecting and it’s actually retargeting, it’s the equivalent of if you told me, Hey everybody that you get signed up for Rise I’m gonna give you $10, William. And I go to the people who are, let’s say that like Rise is a coffee shop actually, and they’re in line at this coffee shop. And I go up to the people that are in line at the coffee shop and I say, Hey, buy this coffee. It’s amazing. Like they’re already in line. And you’re like, well that didn’t add any incremental revenue for me. But now in addition to not adding the incremental revenue, I have to pay you $10 for every one of these people that was already going to buy.
So it decreases that profit then that you made per transaction as opposed to you…

Russell Miller: Decreasing the profit of the people that you were gonna get anyway.

William Harris: Exactly.

Russell Miller: So that’s, that’s huge. If you’re trying to be kind of a data driven business and you know, you know, a ROI driven business, if you can’t kinda accurately think about, you know, what is our tact for bringing a genuinely new customer?
And that gets kind of like muddied, right?

William Harris: So.

Russell Miller: Let’s, let’s move on and think a little bit more about the ecommerce business. So if a customer comes to you like a new, a brand new customer and says, Hey William, I wanna drive more sales through paid media, and you’re going in and you’re kind of looking at that ecommerce business. What are you looking for to see if they’re gonna be scalable?
What are the prerequisites to like doing well with you.

William Harris: Yep. So number one is, is is proven product market fit, so we don’t love coming in where they haven’t actually proven that out yet. We’re, we’re definitely much more on the data side where we wanna,
Is

Russell Miller: there like a certain, like monthly revenue number or, or something, or,

William Harris: No, but let’s say that there’s, at least to the point where they’ve tested things out.
They’re, they’re already getting sales, let’s say organically, but they’ve also tested some paid media out and they’re, they’re at least getting some kind of return. They might not be profitable on that return, but they’re saying, look, when, when we show ads to people who’ve never heard of us, we are getting people to buy.
So there’s product market fit. People are looking for what we want. The price is obviously right enough, like we can, we can work with that. The other big thing is, Yeah, so, and one of the other questions that we ask before we get into actually taking on a client is inventory and, and where they’re at with that, because we’ve run into this before, there was a customer that wanted to double their, their revenue.
And we were on, on pace to double the revenue based on the amount of spend and everything that we were doing. And things were in a really good point. They ran out of inventory. They only ordered 50% more inventory , so we didn’t know that ahead of time. And so the, the reality here that we start to qualify is making sure that there’s also good communication with whoever’s making the purchasing as well.
So that way if, if the marketing team’s got a goal to double revenue, but they’ve only ordered 50% more inventory, then it’s not gonna work and it’s gonna be a failed attempt. We’re gonna end up overspending too much without, you know, overspent now you don’t have anything to sell towards the back end of something.
So that’s the other big thing that we’re looking at.

Russell Miller: That seems like a really key point about, about the inventory sort of matching the goals, especially as we head into like a high spend season where they’re like, okay, we’re gonna have our biggest Black Friday. You may ask them, like, you know, would you be ready for that?

William Harris: Yeah, exactly. And I, I like speaking of Black Friday. Typically what we like to talk about is it’s not the time to test something new, right? You should have tested out whatever type of offer or creative or something that you wanted to throughout the year. And so if you’re wanting to do, you know, a percentage off, you should have, you know, pressure tested that earlier in the year.
See how do your normal customers react? How does that impact your customer acquisition cost, right? In any kind of a way. So you have at least a baseline gauge of that versus maybe something else. If you do like a bogo, buy one, get one off, right? Or if you do some other type of like just a dollar amount off or whatever that offer is, you should have tested a few of those leading up to this to ideally know which one.
And how does that impact your overall profitability? Because again, going back to what we do is on the, on the EBITDA side. Yeah. There’s a lot of times where you can run this and you can destroy your profit. You can, your revenue could be through the. You can acquire a whole lot of best, you know, new customers, but you look at the profit that you made that month and it’s down.
And we’ve seen this before for people too. So making sure that you understand how does that overall impact the profitability of what you’re doing? How much revenue do you have to make up for, to, to make up for the, the discount that you’re giving and things along those lines too.

Russell Miller: Oh, that is huge. I think that’s worth that’s worth sort of getting into on its own.
In other words, you’re making a ton of sales, but they’re actually either, you know, break even or negative. Is this something you’re kind of doing in attribution reporting with a client to be like, you know, work out like, Hey, we can only pay this much for a sale. You know, and you sort of like, you guys agree to that going in something like that?

William Harris: There’s a calculator that we’ve built that basically works out based on different assumptions. So let’s say that we, let’s say that we work out where we do a test of 20% off for a customer. Yeah. We’re gonna look at, well, how does that impact their, their customer acquisition cost? Right? Let’s say that that reduces their customer acquisition cost by 15%.
And then what about their existing customers? What kind of response rate does that have with them? How much what’s the increase in the, the amount that those people are willing to buy? Does that change the average order value at all? Right, so it’s, let’s say that’s, if it’s 20% off and normally your average order value is $200, and now people still get $200 worth of stuff.
Let’s say they still spend $200. They just got, you know, that, that extra there, or was it $200 minus the 20%? And so now the average order value is lower. Cost of goods is sold, is still the same price there though, right? So all of those other things factor in. And so we actually run through all of those different pieces to find out where would that be ideal, and how much more would you have to sell to make up for whatever that loss would be in a profit to understand whether this is gonna make sense or not.
And you might find that, you know, based on those. You’d have to double your sales or something like that. Right. And you say that’s a lot. I don’t know if we’re gonna get that and we can start using this to test these out to see what is the response to the different types of offers, and could you get to a point where you can offset that loss in revenue.
Now the one thing I wanna call out here is that’s just if we’re talking about profit month one or four month zero, however you want to define it. Yeah. The other thing to look at is, is lifetime value. And this is the other big thing that I see a lot of people struggle with is understanding. Are you okay to, to lose money on acquiring that customer on month zero?
If you know that you’re gonna get that paid back in the next two months, then sometimes it makes sense to still do that discount. And the the final nuance there though is to understand to, to acquire more new customers, right? So you acquire 25% more customers that way than you, you would’ve before, and so you’re still gonna get that revenue from them over the next 90 days.
The one thing to call out though is somebody who buys from you the first time via a discount is not necessarily the same type of customer as somebody who bought from you full price. And so if you look at your cohorts, you need to make sure that you segment that out separately to find out do you get the same kind of payback on somebody who comes to you for the first purchase with that or not. You might be banking on revenue on, let’s say your LTV as a whole, but if you don’t segment that out, you, you might be miscalculating how much revenue you’re gonna get from those customers that you acquired during a big sale, like Black Friday or something like that.

Russell Miller: Okay. So what, you know, that seems like a really excellent suggestion, so I, I really wanna call that out for our listeners. If you bring in new customers with a big discount, especially on an event like Black Friday, you know, go ahead and create a segment from them in your, in your analytics tool, your tracking tool, so that you can actually measure their lifetime value going forward and making, you know, seeing what their value is versus another customer that, you know, came in organically and and didn’t buy on a huge sale.

William Harris: Exactly

Russell Miller: Yep. Yeah, that seems, that seems huge. Do, do you find that a lot of your clients. Are a lot of your clients going on that sort of, are they LTV focused? Or is it more that they’re like, I need to be unit profitable on each, you know, on each sale.

William Harris: Yeah. And, and this is you know, which one causes which I guess the chicken and the egg, most of our customers are, but that’s because we maybe attract customers who are focused that way and we only take on the customers who are thinking that way, if that makes sense.

It’s kind a prerequisite before we even take somebody on is talking through our approach and making sure that they align with it. So most are LTV.

Russell Miller: Absolutely. Is that because they’re like, maybe they, they just have a longer time horizon and they’re venture backed or, or? You know, something you coach them into?

William Harris: A little bit of both. I would say that you know, PE and VC backing certainly makes a big difference. But I would say that that just allows for a longer length of time. Maybe you go from, you know, one or two months to three or four or five months right? Or whatever that number needs to be. The, the big thing that I’ll call out there is, that was true during a, a really good economy.
You need to look at them very differently right now. And, and you don’t want to be as aggressive with that as you maybe would be during, you know, a really exceptionally growth economy.

Russell Miller: Right. So we are, you know, understanding that we’re all, you know, the, the dollars are maybe pulling back a little bit and, and will be a little bit more conservative.
So looking for like maybe a three month payback, four month payback versus like five, six, seven.

William Harris: Exactly. Yep. And you have to figure out what that is based on, you know, what your overall anticipated goals are. Right. Each person. But yeah, looking at how do you, how do you adjust for that?

Russell Miller: Looking at the spectrum of ad channels in the last couple of years for consumer businesses, we’ve seen several new players come online.
Maybe the biggest one has been TikTok. In the channels that, you know, but we all, we still have the, the the major players of like Facebook and Google. What’s working well for your clients right now?

William Harris: So Facebook absolutely still the best acquiring new customers by far. The, the ad algorithm there is, is significantly more mature.
The, the reason why people still under invest, I think on Facebook right now and why it’s gotten beat up so much is because of all of the iOS 14 stuff and, and things like that. And so people weren’t tracking as much, and so they just saw what they thought was lost revenue, which, which wasn’t.
And, and we talk about the, the ROAZ Death Spiral, which is an article that we’ve got on our, our blog, which walks people through what actually happened. Again, if you look,

Russell Miller: This is a great article by the way. I just wanna call out to the listeners. Check that out.

William Harris: Yeah, it is a long one. It is 7,000 words. Fair warning to those who do go down that path, but it is very indepth. And so there, there’s usually Facebook’s performing significantly better than people realize it’s performing.
So if they can get to the point where they understand attribution better, it is still the best one and the most mature and the most stable of all the ones that you’re looking at for that. Google Ads is, is by far the best for capturing existing demands. So those two, if you’re only going to do two, those are still the ones that we 100% recommend scaling.
They have the most stability you actually can scale and reliability. Between that though, YouTube and TikTok are, are my tie for third place, because they’re both doing very well. YouTube both of them, the, the interesting thing is they both need very different content than what you’re used to producing.
YouTube you do need a lot different content than what you’re putting on Facebook. And TikTok, it’s very different. TikTok content works great on Facebook ads. Not the flip, not the reverse. You can’t take what you’re doing on Facebook, throw it on TikTok and expect much out it. It’s, it won’t perform very well. And so that’s why I think a lot of people have seen a lot of problems with TikTok.
It’s, it’s shiny, it’s exciting. They test it out. They don’t get the performance they want because they don’t have the right content. And the biggest thing there is recognizing that you probably don’t know what is good content there. You need to just work with somebody who does and usually, Yeah, it’s going to look really bad to you and you have to get over it from a brand standards kind of perspective.
And just say, okay, I trust you. Let’s get this in here and let’s see what happens. Cuz it usually looks horrible in my opinion. But that’s cause I didn’t grow up with TikTok, I grew up with Facebook.

Russell Miller: Yeah. So it’s something to maybe like, experiment with and understand that the creative is gonna be, probably amateurish compared to what you’d see on, on Facebook.
Let me ask you for your hot take on a couple channels cause I think this will be fun.

William Harris: Go for it. Spotify. Mm. Okay. So I, I, I do like Spotify. One of the things that I like about it is that it’s like auditory type content as well.
Right. And I think that we, we, we miss out on how impactful audio can be. That said, if you’re not quite at the point where you’re ready to invest into Spotify or other programmatic channels, which was, I had number four on my list. Yeah. So I’m glad that you got to this. Yeah. If you’re not there, YouTube has YouTube Radio.
So if you’re already on Google, you can get on YouTube radio and you can at least start testing out a concept of radio without having to necessarily say, wait, how do we get into programmatic? And that’s just a way to say, well, how about let’s test some audio type ads. I think it’s a great, it’s a great thing to add.
It won’t perform as well from a direct response perspective. And so what we usually say is before you start getting into, let’s say, a lot of the other programmatic platforms, right? I like to see brands spending at least a hundred thousand a month, if not 250,000 a month on the other platforms. Get to a point where you’ve plateaued there and you’re saying, all right, let’s go to what’s next and what’s incremental here.

Russell Miller: Okay, got it. So that, that sort of answers my question. Would that include programmatic too? Like get to a hundred, 200 K on Facebook and Google before? And for our listeners who don’t know, programmatic is just sort of the new name for display ads. So, so get to like 100, 200 k in, in advertising there before kind of messing with the display world.

William Harris: Yeah. Yeah. Not always. But, but most of the time, yes. And the, and the reason why is because those other two are so, so tuned into, so, alright, and here’s why. If I’ve got a couple more minutes to go over this. Yeah. Here’s why this makes such a big difference. You have to think about all of the websites that have the Facebook pixel and the Google Pixel installed on them from a tracking perspective.
Think about how many years they have been collecting data on literally just about every single website that exists on the planet, right? Think about how they have used that to understand people’s behaviors in such a big way. They still do the best at driving those results. And then if you get into, let’s say Facebook and in TikTok as opposed to programmatic from a direct response perspective. If you see an ad on either one of these platforms, they take up your whole screen. You can’t help but see. You have to move past it. In programmatic, and I’m sure you’ve experienced this in a lot of programmatic, in a lot of display type ads you, you can be served, you know, 50 different impressions on the same page without your, but it’s in your periphery.
Right? Right. That’s so good. It’s still good, but it’s not necessarily usually going to have as good of a direct response impact. And so if you’re saying, I need direct response right now cuz we’re spending under a hundred thousand dollars, then you probably need like, more immediate impact from your dollars.
You’re not at the point where you say, Hey, I need to broaden my, my expanse and my reach. That’s not to say that programmatic does not have its place where it can do better, but it’s seldom and it’s rare. Usually in those cases it’s when you’re struggling to even get anything going well on one of the social channels because you’re in a touchy subject like, you know CBD or Crypto or something.
Yeah, yeah. Something else where you’re like, it’s a lot easier to go there. And they have some dedicated programmatic partners that are really good at having already the best spots and placements on some of those you know, unique channels.

Russell Miller: So, okay. One more, one more hot take. Twitter.

William Harris: Okay. I’m indifferent on Twitter right now, and I’ll say that I, I, I, I don’t know.
We haven’t found anything amazing. We don’t have any runaway success on Twitter, but I’d say that I’d almost lump it in with the programmatic, where it is still a very good channel to use. There are some, there are some different targeting options you can do on Twitter that you can’t do elsewhere. That I think just make it interesting.
I think it can be, let’s, let’s throw that in with programmatic and let’s throw that in with Microsoft ads, meaning Yeah. Is a good channel to add on when you’ve kind of plateaued on some of the other channels and you want to reach a little bit differently, in a little different way. The creative that you need there doesn’t need to be dynamic in some way where it’s like, look, if you’re not a very creative focused company like TikTok is hard, but you can throw Twitter some stuff on Twitter and, and get some decent performance without having to create like wildly new creative.

Russell Miller: Okay, so like lukewarm I, I personally I’m, I’m not optimistic given recent events. So I’ll, I’ll leave that there. I, I always love to ask founders, especially after they’ve been in trenches for a while about some of their war stories where brands where they’ve just really killed it.
You know, to get a chance to kind of show off and, and tell you about like some of the best work you’ve done. Are there, are there any are there any great war stories you have that you’d like to share with us today?

William Harris: There are two that come to mind. Recently there was a client that just came to us and without changing any of their creative, well, and I’ll even say this is a client that they were our client for about maybe three years or so.
They got a consultant to come in and the consultant felt that we were, were too expensive of an agency and that they wanted to go with a, a, you know, a different agency. That lasted two months and then they were back in and within without changing any of the creative, without changing any of the budget.
The only thing that we did differently was we set up things within the campaigns correctly. What we would say is like the boring technical stuff that everybody else seems to say doesn’t matter. Just let Facebook do it all for you. They went from, don’t quote me on the exact numbers, but something about like A two X MER or ROAZ if you would.
So a five X MER or ROAZ. Literally the same day. It sustained. And, and the only thing that was different was just the technical stuff that we set up. So that’s the one that I really like.
Because

Russell Miller: they do say, you know, I think the party line from Facebook is you know, we have machine learning. You know, throw, just give us a box with all your creative and, and, and we’ll take it from there.
Is that how you guys respond? Is that sort of the, you know, the approach, you know, you take or, or is it, or you like, sort of against it?

William Harris: Facebook is brilliant and this is why we still say that it’s the best for going after this. But Facebook is limited. Now there are ways you can make it a little bit less limited, but they’re limited in what they know about the business.
So here’s the, a couple of examples. Let’s say that we decide that and this is a really dumb example, but let’s just say that we decide that on average people named Russell end up spending three times more within the first 90 days than people named William. I’m gonna target a whole lot of people named Russell then, because even if it costs more for me to acquire Russell than it does for me to acquire a William I’m gonna get more overall revenues in a pretty significant way that’s gonna impact my profit a lot more.
The problem is that Facebook doesn’t know the difference between that. So you optimize it for a purchase and it’s just gonna go for the lowest cost purchase, and maybe you can acquire William purchase for $5 less than a Russell purchase. So that’s why I say that like it is, it’s really good.
Facebook is really good if you set it up correctly. And I think that most people just, they’re lacking that understanding of, of like the nuanced things we talked about there from like really making sure that it’s actually prospecting, but then Facebook just doesn’t know what it doesn’t know yet, but there’s ways to feed it a lot more of that information, which we do.
We do this on Google too, where we’re, we’re actually feeding Google profit for product and stuff along those lines, so it’s a much better it. We’re enabling the algorithms that are very powerful to use better data to optimize around what we actually want.

Russell Miller: So it sounds like, you know, one of the, one of, you know, for this customer, you helped out, one of the things you brought in was, was really technical setup and also maybe, you know, using their first party data, you know, to really figure out an edge and like who to target. Is that fair?

William Harris: Absolutely. The other war story then is when. We just published this one on our blog. There’s a 50 million Shopify store. They were seeing a massive decline in customer growth year over year. And we noticed that this was actually related to, again, a poor understanding of the boring technical stuff.
So I wanna say that they were down. They, because of the size, they were down 9% customers year over year month over month, over month. I wanna say it was like maybe four or five months in a row prior to coming to us. Within one month when we flipped on our campaigns. Again, no new creative, no new budget. This was just flipping things around. Actually I think there may have been a little bit more budget at this point in time. Just as we were ramping up for them throughout the course of the month. Increased the new customers by 29% year over year. So this was a, they, they were having a really good, down trend and we, we were able to flip that around.
And then when you look at, I don’t know if you’re familiar with Veros. Veros is a really nice benchmarking company that we’ve been really enjoying using. It is Y Combinator backed and they have they have performance lines where they’re looking at quartiles. So they’re gonna bring in a bunch of other Shopify stores and, and instead of it just being irrelevant benchmarks, which I’m, I’m very much against, they try to get a lot more relevant.
So they’ll say, okay, how about companies that are within the kids and baby’s vertical space that are apparel, that are spending a hundred thousand dollars a month at least, that have an average order value of X, Y, Z. So now that’s a much more relevant benchmark. And within this benchmark data, they’re looking at the quartiles, the bottom, the median, and the top core tile.
And then view, right, so there’s a line of view. We actually destroyed their chart with this customer. Because there’s the, the, the bottom quartile, the median the top quartile, and then us. And so we’re so far beyond that that you almost can’t even see the detail of the other three lines that are in there.
And then view, right, so there’s a line of view. We actually destroyed their chart with this customer. Because there’s the, the, the bottom quartile, the median the top quartile, and then us. And so we’re so far beyond that that you almost can’t even see the detail of the other three lines that are in there.
ROAZ of 6.28 compared to the benchmark of 1.48 over, over this. And so I think it’s, again, that was another good war story of this was a, a significant difference. And it’s almost overnight.

Russell Miller: And is this, and, and is this, when you, you know, you said like paying attention to the boring technical stuff. When you came into this account, you know, what, what were you seeing? Like it was just set up as sort of one undifferentiated blob campaign or?

William Harris: No, it was set up very well. I would say that the company that was doing this stuff not negligent if that makes sense. Right. Companies, I think they were, they were absolutely doing a good job and was set up with what you would say is best practice if you follow all the Twitter threads. Right. But, but this comes down to, again, I think the majority of people out there, and actually the Veros founder said this to us when he asked, what are you guys doing that’s so different from everybody else? And I explained some of these concepts to him.
He said, that’s interesting. You the only one that said that. Everybody else has said that it’s about the creative. And so I think that it’s just that there’s, there’s maybe just a lack of looking deeper at the setup and, and how impactful it can be. And again, going back to those exclusions, it makes a big difference.
Another thing that we really push for is instead of allowing the algorithm, you can tell it to optimize around a, a seven day click, one day view. We force it to only optimize around the seven day click. The, the benefit there is you’re saying to Facebook, you’re saying, I don’t, I will not let you take credit for it unless you caused that person to click and convert that person.
The downside to that, and I think why some people end up not doing that is cuz they want more data to flow through, right? Right. They want, well I want more purchases so that way the algorithm can get smarter. The problem with that though is that you’re allowing too many view through conversions that Facebook didn’t necessarily cause. You’re allowing that data into it so now you just have, let’s say, diluted data instead of very specific, intentional, incremental data. So we still use the view throughs in reporting to understand how that’s happening. You’re still gonna get more view throughs when you force it to go after the people that are going to click and convert. The difference is you’ve just forced the algorithm to have to work a little harder to find those right people.

Russell Miller: So you’re opt… yeah. So, so for a listeners, you’re actually, you know, there are sort of broadly two ways of tracking ads. Does someone see an ad, you know, that’s called an impression or does someone actually click on an ad? So that’s a click through. Facebook has a initial optimization window of, well, we can do a click or a view within seven days, but those are not the same, where a click is a very intentional sort of thing.
Whereas view through, they may or may not have have seen the ad. And so if you’re getting, if you’re just focusing on the click through, it’s a much, it’s you know, arguably better data.

William Harris: Yeah, exactly. And so by doing things like that, and that’s just, you know, two of the few things we mentioned about sending through, you know, profit into Google, there’s, there’s just a, a, a number of things that we’re doing from a technical side of things that I think it’s easy for people to overlook and, and just say, oh, your creative’s just not good enough, which is also always true.
So we don’t disagree with that. We’re just saying that doesn’t have to be the only thing that you go to. There’s a lot that can be done without the new creative. And then if you also had good new creative on top of that, it’s, it’s, you know, exceptional.

Russell Miller: So, yeah. So maybe, maybe the lesson there is, you know, creative because it’s visible, you know, is what people usually reach for because it’s sort of obvious. But there are a lot of, you know, subtle things to look for in the setup that it can actually drive returns a lot more that people can dig into.

William Harris: Yeah. Well said.

Russell Miller: So I wanted to also, we’re sort on the down end. We have like 10, 15 minutes left. What element you know, you guys are primarily focused on advertising. Does email or email capture or email retargeting, you know, how do you think about that in terms of like your overall like ad profitability and ad ROAZ. How does that plan.

William Harris: So I got another hot take on this one. Hopefully that’s okay. Right. Contrarian point of view. There’s, there’s, there’s several things that I like about it and several things I strongly dislike about it.
The one that I really don’t like is, I’m not a big fan of from an ad perspective, sending people to, and I’ve seen people do this and even suggest this client suggested too, right? Of, hey, let’s optimize for an email capture from an advertising perspective. Again, I’m talking about ecommerce.
I’d say optimizing for an email capture if you’re doing lead gen a hundred percent, but we’re talking ecommerce. More often than not you, you’re gonna be better off just optimizing for the person that is going to to actually buy. And, and part of this comes down to, if you think about like those fishbowls that you sometimes see at a place where it’s like, Hey, drop your business card in to win an iPad.
How relevant are the majority of those business cards in there for that business? It’s, it’s very few of those are actually interested in whatever that product is that that company is selling that has the fishbowl out there. Right. The same thing happens if you optimize for, for just email capture.
From an ad perspective, you end up with a lot of emails that are not even remotely qualified buyers, maybe don’t even buy things online at all which is part of what the algorithm’s doing when you say optimize for purchase, it’s looking at it saying, Hey, you know, my mom, Deanna, she doesn’t buy anything online ever, so I’m not gonna optimize her.
She might throw her email into something. Right. So, okay, great. You said, well, she’s gotten zero chance of becoming your customer. So I’m not a big fan of doing it from that perspective. I am a big fan of doing email capture. When you say you’ve got somebody to the website, and they didn’t purchase and they’re about to leave. Exit prompt, I’m absolutely in favor of, they’re about to leave and there’s a good chance you might not see them back there.
There’s a lot of benefit to that. The biggest thing that I would say though is in all of these situations, these are different email addresses than your customer email address list. And I see too many people who do their emails where they’ll send them and they will lump them all together in a lot of the things that they’re doing instead of segmenting out those.
And you have to remember that if your email list is currently, let’s say, you know, a hundred percent customers. And then all of a sudden, after doing a lot of this stuff, your email list is 50% people who have bought from you before and 50% people who haven’t, and you treat them as one same list. Your deliverability is going to go down because you’ve got less people that are interested, though they’re not as interested as the people who have bought from you before. Your clickthrough rates go down. It can absolutely ruin your ability to actually reach your customers who have bought from you before too. So if you,

Russell Miller: So you definitely wanna segment those out. You wanna have like your, your list of your, your customer, your customer list of people who’ve actually purchased from you, maintain that as a separate list versus people who have entered their email into a popup because you know, they land on your website somehow.

William Harris: Same thing that we do on the ad side. Look at it as two different funnels. There’s like a circle or a cyclone almost, if you would, of like, awareness, consideration, decision. Awareness, consideration, decision. Up until they make a purchase. Once they do, they’re in a completely different set of things and the same thing from an email perspective.

Russell Miller: I wanted to ask about kind of the, the biggest change in the past year or so, and this is just you know, you know, could be a trend, something that you’ve seen in, in ecommerce or something you’ve seen in kind of, you know, the broader macro economy and, you know, how, how can you help your, your customers deal with it.

William Harris: : Yeah. And I think we touched on this a little bit before, which I, I would like to say that we’re done with iOS 14 issues. Yeah. But at least 70% of the accounts that we take on haven’t been set up properly. And so that’s still a big issue. But the biggest change is obviously the, the, the macro economy.
And so the, they, let’s say the, the almost inevitable recession that, that we’re heading into. And, and this has reduced a lot of just overall, buyer demand in, in a number of places. Now we’ve got plenty of customers, clients that are still up year over year. But there’s a lot of, especially if you’re a bigger client, if you’re a bigger customer, if you already have a significant market share, you end up just kind of riding the wave a little bit of the overall market demand. There’s, there’s less market capitalization that you can do. Right. And so for some bigger customers one of the big things we’re looking at and for smaller ones who are maybe experiencing this too, is really trying to think through what is their overall, again, EBITDA growth goal for this year.
And, and, and let’s say that there’s x amount of inventory, maybe they already ordered inventory ahead of time before coming to us thinking that this was gonna be a bigger year than 2021 for them. But they have been down every single month, year of year or, or they’re not growing the way they thought they were.
Now it becomes more of a conversation about how do we adjust to maximize the EBITDA that you have with the products that you, with the inventory you have on hand. And one of these is you can, you can advertise more aggressively to sell more product, to move more product, which is good from a cash flow perspective.
And there are some clients that opt for that route where they say, look, we’re gonna go ahead and discount stuff. We’ll take that hit on that side because we need the cash flow. Or you say, no, we want to focus on EBITDA. It’s okay if we don’t sell as high a volume of products. We still wanna focus on just making sure that we’re profitable on what we do sell.
So the biggest change I think, is just looking at that with clients and helping them to understand and forecast better with everything that’s been going on.

Russell Miller: So it sounds like having really, you know, given that we’re heading into a recession, having an honest conversation with your clients about what their goals are and what their trade offs. Are they going for, you know, new customer acquisition or is it really more about, you know, profit per sale and being very clear on that and then being very clear on matching the demand generation side with, you know, what the actual inventory is. Even now, two years plus pandemic, we’re still working through a lot of China supply chain issues, probably for a lot of your clients. And you know, the demand that you are able to achieve for them may not match the inventory they’re, you know, they can actually get.

William Harris: Yeah. And a good way to think about some of this too is really like going to Google trends and just seeing demand for like, let’s just say like big giant things. Overall. So if it’s, let’s say that you, it’s coffee, then it’s just like, just literally just typing in coffee versus like something very specific, right? Like what’s the overall demand for people searching for coffee? And, and let’s just imagine that the demand is down. If you look at, let’s say November versus November this year for people searching for coffee, and I, I didn’t look this one up ahead of time, so I dunno, but let’s just say that it’s down 40% year over year in the number of people searching for it.
Well, let’s say that last year you were in a growth mode. So you were being aggressive, you were already working towards, let’s say, lifetime value. So you were going for the highest CAC possible as you were trying to really push and go after everything you could, right. You were willing to to to be aggressive on that.
And let’s say that now, if you want to still achieve that same amount, you can still take it, but you’re taking market share from somebody else, but the overall demand has gone down for that. Even if you’re still relatively small, if you are already maxed at your efficiency level, then you have to understand that like continuing to grow while the market demand is going down means that you are going to have resistance.
Even if you’re small, there’s still resistance against where you were. So if you were seeing, let’s say, a CAC of $20, if you want to grow, you might see in everything the van is down, you might see CAC of $25. And you have to determine is that worth it for you at this point in time or not, or is it worth it for you instead to kind of stay, stay at course, pull back, maybe just go for a CAC of, you know, $18 or something along those lines.
And if you do, how much lower overall revenue does that make for you? Cuz the other thing to remember is overhead. Yeah. And so if this impacts your overhead where you say, I actually need X amounts of sales, and that’s where we kind of just work with customers to help them do a better job of just kind of helping them navigate through all of that. Like where is the sweet spot?

Russell Miller: Yeah. So really taking a holistic view so that if demand is, is shrinking, we might have to fight much harder for that incremental growth because there’s just less pie to go around, you know? Are we willing to increase our CAC to go after it? Or maybe we say, oh, you know what, beyond a certain level, it’s not worth it for us.
So understanding what those drivers are for the company, and you and Elman are kind of helping, you know, steer your client to kind of make those trade offs.

William Harris: Exactly. To at least help open up those conversations so that way Oh, you’re right. Like what? The two ends of the same string. Which direction do you wanna pull it? Yeah. And we can help support either goal.

Russell Miller: The the stupid version of that I saw was people thinking that the sales from 2021 were not a windfall and were just like a new normal. And that would continue.
This has been really fantastic. I wanted to end with just asking you for a little bit about your backstory and the backstory of Elumynt, because it is you know, super exciting to be, to become, you know on these, on these Inc. 500 lists and one that’s in Minnesota. How did that, what’s sort of the growth story there?

William Harris: Alright. Best for last, I guess. Huh? There’s, so, there’s a really long version of this that starts with the rock band that I was in during high school. But if we fast forward some, some interesting career moves. Yeah.
I was actually leading growth for a, a fun VC backed SaaS company called When I Work. There’s amazing founder there, Chad Halverson. I got to shake Alan Patricof’s hand a few times. If you don’t know Alan Patricof for the audience there. He’s like the godfather of VC, like just a really cool, you know, old dude now.
But it was like, it was a big deal for me. And basically with the help of our entire team there, we were able to grow, when I worked that company, by about 270%. In the first seven months that I came on board. So exceptional growth. Ended up getting poached by a really interesting RC car and drone company website called Dollar Hobbies. And they were crushing it. And we expanded them from, I wanna say about 3000 products to 70,000 products in the course of a year. And that was my entry into ecommerce.
And so I remember applying a lot of the same concepts that we used in SaaS into ecommerce, and I was surprised that more people weren’t approaching ecommerce this way. But keeping this as short as possible, fast forward a little more, and I start writing about this stuff that we’re doing on a different blogs like Entrepreneur, Fast Company, et cetera.
And before long I had a lot of people just reaching out saying, Hey, I like what you’re talking about. Can you grow my business too? And that’s when I launched the agency. And at the time, I didn’t think agencies needed to be disrupted. I had never worked at an agency. But I quickly found out that most of the talent at typical agencies didn’t understand the deeper aspects of business.
They were looking at things like ROAZ instead of profit. And while it seems like it would be the same, obviously I could assure that it’s very possible to increase the reported return on ad spend without increasing the EBITDA. Finance team sees that, et cetera.
So it wasn’t the agency’s fault though. Most of the marketing teams at companies they just don’t know what profit looked like at their company cause they weren’t even privy to the P&Ls. So, so anyway, so we, we, we go through this and that’s kind of what started down this path of, okay, how do we continue to find an agency that can operate around what that business’s actual goals are.
And that’s kind of what started the, the successful launch of this. Where all of our growth has been through referrals and a little bit of content marketing through things like being on a podcast or writing on Entrepreneur. But this is just doing really great work and people saying, wow, you guys get it differently than the people we’ve talked to before.

Russell Miller: I think having that business mindset is, you know, really huge and, yeah, really a differentiator. I, I think there a lot of agency people are kind of like, you know, they see the Facebook dashboard and that’s sort of as, as far as it goes. So that is, I think, a really interesting point of differentiation for you guys in Elumynt.
So we will, we will wrap it there. I want to thank our guest William Harris: of Elumynt. Super fast, high growth ecommerce agency. If, if you need to blow your sales out, these are really good guys to talk to. So thank you so much, William.

William Harris: Thanks Russell.

Russell Miller: Thanks for tuning in. This has been the Heroes of Ecommerce Podcast, brought to you by Ryzeo, a proven email system that’s made millions for ecommerce sites. Be sure to subscribe to our podcast so you don’t miss an episode. For even more insights, visit our website at ryzeo.com where we share resources on email marketing that grows sales. Until next time, happy marketing.