Justin Miller 06:04
Yeah, it’s a good analogy. And the book’s Jurassic Marketing and quite possibly next time you see me speak Profit 911 maybe a little tiny thing in Jurassic Marketing front and center. I won’t give you the long version of the bio, so we’ll blow through a lot of stuff that normally I spend some time on, but I’m an entrepreneur at heart. Most of what I did was learning for my own companies, starting at the age of 14 as a mobile disc jockey doing weddings. Shortly after I got out of the wedding industry, I consulted to that industry first, but we ran a digital marketing agency and my bread and butter was conditional follow up multi-channel builds in Infusionsoft at the time. So that’s the world I came from. That doesn’t exist here at Profit 911 anymore. We are 100% a direct mail provider at this point. So we have devolved, which is kind of funny, and we continue to devolve even more and our results continue to get better for us and our clients. So obviously all media still works. We’ve chosen to go backwards in time as far as what we provide and what we do in our team meetings. We also like to have fun. So there’s dinosaurs everywhere. And you know, if you happen to become a client with us, I’ll give away a secret that you’ll get a chocolate dinosaur and all kinds of fun stuff. But we physically print and mail millions of pieces a year, literally, in this building, about 50ft that way. So we are hands on from concept all the way out the door for a lot of people. So we get a lot of insight into what’s going on, and I will share as many examples as I can. Not naming most of the clients, but I’ll share what I can. And I came to this world from Dave Dee, who I see the group and some of you know and are still connected with. I saw him speak about direct response. He was the first person to ever bring up the concept. I don’t believe he called it that, but that was before I learned how to drive a car, and it kind of set me on this path, which eventually led me to Mr. Dan Kennedy, who many of you also know, no doubt. And then eventually, as Brian said, I found my way over to Joe Polish, and that’s where we kind of hanged out, hung out a little bit, and I’ll play what he has to say.
Video plays
Joe Polish 08:17
Love to hear what you have to share and I want to also encourage people. There are very few people that have the depth of knowledge and that can actually help you do direct mail is this guy. It is a dying art in terms of some of the best direct mail experts are old. A lot of them just give ideas. He actually has a whole business that can do it for you and stuff. And so I’ve said this many times, but I haven’t put much emphasis on it at this meeting and in previous meetings is, we have seen people that if you just add direct mail to the mix of your marketing, oftentimes it could double or triple your business. And this is a guy that can help you do that. So at some point, if you don’t know him, get to know him because he and I don’t even like saying this many nice things about him because he’s a really bad person, but he does have good comments. No, I’m just kidding. So but yeah, he really does know his shit and he can really help you out. And so I would encourage you to get to know him if you haven’t already.
Justin Miller 09:13
And if you don’t know Joe, he can’t give a compliment without a quip. So I don’t think he thinks I’m a bad guy. But we have a weird sense of humor in that room for sure. As I’m sure Brian can attest to. All right. So typically, my entire presentation is convincing people that direct mail is not dead. We’re going to do that with just this one line right here. So there you go. If you go see me speak to any other group, that’s not like I haven’t been prepped. They’re astute marketers. It’s a whole thing to ultimately lead them to come to the conclusion direct mail is not dead. So you guys already know that. I’ll go into some of the psychology behind our philosophies today, the techniques we’re using. But be very, very careful, especially those of you that said you’re doing client work that the vast majority of the world still believes it’s dead, and probably more so by the year. So don’t ever forget the importance of this. Don’t ever forget the hurdle of overcoming it if you’re trying to add marketing to someone’s mix. They think it’s dead. On the surface, it would appear that way to most people, and it’s a hurdle to overcome. A challenging one as a provider, by the way. But today I don’t have dinosaur references. I have football references, which is going to be interesting with dinosaur backgrounds, hence the excuse some of the polish on the presentation today. So yesterday, if you’re an NFL fan, particularly if you’re in Illinois like me, this is a rendering of the Bears new stadium, that $5 billion stadium next door to Soldier Field’s current location. So when I say football, the vast majority of you think of American football. This is the Emirates Stadium in London, where the Arsenal Football Club plays. As far as the world goes, this is football. So I want to kind of use this analogy for direct mail, because particularly astute marketers, when we’re talking about direct mail and learning things, we are learning about, you know, American football, we’re in the trenches, large quantity, split test, everything like that that’s of extreme interest to us marketers and we hear about. The problem is it’s a very, very small minority of people using direct mail that are mailing significant quantities of it. So the vast majority of people are not mailing enough quantities to use techniques or learning about. And they’re playing soccer and there’s money in soccer. And most of our clients are in soccer. And I will share the examples from them where you might not be able to use everything that you want to. So we’re going to talk about what to do when your results are ambiguous or statistically insignificant, which is the case for a lot of businesses that are or currently could be using direct mail by running low quantity campaigns. So high value, low quantity. Interesting place to play because it’s a little bit different angle than you’re typically used to. So in case you’re still stuck thinking that American football wins. So here’s some stats. Super Bowl ‘22 150 million viewers. There’s some of the World Cup finals. You know, triple the viewers. Another one, so 2022 Super Bowl versus World Cup. So US viewership, Super Bowl 113 million, soccer 25 million. That’s why we think of NFL. Global reach 40 million versus 1.5 billion. So needless to say, if we apply this to marketing, you know, the vast world that could be using direct mail is not, because they’re using the wrong game or have never considered the right game. So let’s dig in a little bit. So what if we’re going to live in a world where you can’t split test? If you analyze the numbers too deep, you come to incorrect conclusions. What if there’s no existing control that you’ve used or you can find anywhere? And what if you only need one tenth of 1% response to attain success? A lot of times we hear 1 and 2% as a standard number. Obviously, numbers are all over the place. But what if one tenth works for you? So this is the world I spend most of my time in. Of course, we like large campaigns too, by the way, but they just are not the majority of our work. So, Brian, in your book, Overdeliver, you have a concept of pyramiding, which has to do with testing and audience sizes. And we have this conversation a lot around the office. You know, when things clients come in and want to do testing. Your examples in the book, the first test of a list is 5000, so the top, the small part of the pyramid. Then it’s hey, let’s see what happens when we move to 25,000. And then if that works, let’s see what moves to 100,000. Challenge for me and a lot of others is what if the entire prospect universe is 10,000? What if it’s only 1000? Or worse yet, what if it’s only 100 people prospects, not leads? This is the direct mail use case for a lot of business owners. Or it’s the direct use case for a lot of businesses that could be using direct mail but don’t know how. Or maybe they dip their toes in and were disappointed by results because they were looking a little the wrong way. And I’ll give you the earliest example I have, which would be, you know, my own, which was the wedding company. So we were, you know, a tiny local wedding service provider serving high end weddings. Our metro area is about 400,000 people in any given year. There were about 1500 to 2000 people having wedding receptions that may be hiring our type of service. That was our whole universe. The fraction that would pay for our service was even less. But that was it. That’s all we had to work with without changing the rules. And those were the rules as we set them. That’s where we wanted to play. And, you know, we made direct mail work, but we couldn’t use a lot of the techniques and iterations that might be appropriate elsewhere. So can you still win? Can you still win with direct mail when all that stuff stacked against you? Yes, of course you can. However, you got to play soccer, not football. So let’s talk about the rules of soccer. First, testing results are not statistically significant. We’ll dig into each of these in a little more detail. Second, no real split testing. Third, the universe is finite. Fourth, past data is limited. And five defining success is a little different. So like the internal marketing heart of some of you just shuddered. I could see it like this doesn’t usually sit well if you’re a real marketer. So let’s talk about it. And by the way, this is not an excuse for laziness. So we’ll talk about when to apply these rules and when not to. So they’re not across the board. They’re specific to these types of cases. But testing results are not statistically significant. So we have a lot of people probably not a day that I don’t overhear a campaign manager here talking to a client about some sort of test, which is good, of course. That should be part of it. However, most of them had no idea what that really means, much less have ever pulled up an a b test calculator, which is so easy to do with Google these days. Or their test involves their individual piece. So obviously people seed mailing lists and once or twice a year we get a client who put their own address in the mailing list and they didn’t get their piece. And holy crap, you know, trust is lost because they didn’t get their piece. And I have to look around and say, okay, well, that’s not statistically significant. It went out the door. We can track the rest. But, you know, this is the rules. So AB test calculator is very easy to find, very easy to use. This is one I don’t even remember where I stole these screenshots from, but I ran these tests today. So this is if we split test a thousand pieces, which for a small business might be a realistic mailing size. So we send variant A: 500, variant two: 500. We get two responses: version A. One response version B. And you can select your confidence interval there. I went as lenient as possible at 90%. Guess what? Result not significant. Rookies to direct mail or maybe your client might take that as significant. This one outperformed the other. Let’s stop using the other. Happens all the time. I kept bumping till we got to significant with this small sample size, and we had to get to six conversions versus two. So technically now we’re at 90% confidence. Now the stat says that it’s still a damn small test. So I believe it more, but not significantly. So what on earth do we do when that’s the case? You just have to realize your results might not be significant. You have to be careful with what you do with that data. So some elements are still going to be guesswork, some is going to be gut, and hopefully it’s experienced gut. This is the value you’re providing, is experienced gut if you’re working for clients. And some of it obviously is previous experience. But if we’re working with small enough numbers like this, you know, we have to play with gambling money. So what is the investment versus the possible reward? Oftentimes the risk is a lot more minimal. You know, the test wouldn’t be justified if you could even do them. Oftentimes the split test is not even justified screwing with because the list is so small. So we make sure that if our clients are in this world where their numbers are small enough, that they understand what’s really going on here, and that there is more of an element of gambling than we traditionally would want as a student marketers. But that’s not to say it’s wrong. It’s just to say you have to understand when this is the game you’re playing. Number two, no real split testing. So as you can see, depending on the confidence interval you want, you have to bump numbers. You know that, that’s basic statistic stuff. Sometimes though this becomes interesting because you do have to iterate. You don’t want to get incredibly lazy. You do want to try and beat your control, even if you know it may not be significant. You want to see bumps in numbers. Sometimes, though, you got to throw out the baby with the bathwater. You know you always hear, don’t throw out the baby with the bath water. Sometimes you have to. Sometimes in this world of small numbers, you have to change everything. You have to change multiple variables at once. Like this is against like every rule of market testing, right? You have to learn to live with acceptable error. So is this a big deal if the assumption is wrong and we go forward? You have to actually think about that and move on with it. And sometimes you have to be okay with just knowing something works and not why it works. So we everyone should have these pieces, be it online, digital or offline. But you test multiple things and you just have no damn clue why this one’s outpolling the other. So, you know, we have split tested things that were right on that verge of statistical significance and done 5 or 6 iterations and got minor results. And eventually on the next test, we’re like, all right, screw it. Let’s do something drastically different and see what happens. And we did and just blew the first one out of the water. And now we don’t know why because we changed so much. But guess what. It doesn’t matter. We have a higher baseline to start from now. And now we can test from that. So sometimes that happens. And it’s more so in small number world. Or maybe you run a test and both work. Maybe they’re different. Maybe they different prospects are responding to different things and it doesn’t matter if you improve the other one. Prospect A was going to respond to that style and prospect B the other, and that’s okay too. So oftentimes with smaller quantity, we’re looking at, you know, ten ways to convert one prospect instead of one way to get ten. Obviously the numbers are expanded a little. It’s not literally that. But that’s the general concept is we might have to do more things in more ways. It was really funny in that wedding business as well, that, you know, we were a high end provider and being from Dan Kennedy marketing world, you know, we’re ugly marketing dictates. We ran some and we ran it against more traditional, you know, glossy, image oriented. And guess what? They both worked. They both worked. And I don’t think it was the same consumer that would have responded to the other piece versus the other. So we ran both simultaneously. People probably didn’t even realize it was the same damn company, but it worked fine. So you have to be really careful with what you do testing wise and understand that.
Outro 22:51
Thanks for listening to the Timeless Marketing Podcast with Brian Kurtz. Visit BrianKurtz.net and click Podcast at the top of the page for a full transcript and show notes. If you are interested in working with Brian personally inside of Titans Xcelerator, go to briankurtz.net/help to see how Titans can help you grow and scale your business. That’s B-R-I-A-N-K-U-R-T-Z [dot] net [forward slash] help.