Ralph Burns: AI and the Future of Digital Marketing

Introduction

[00:00: 32] Sam: My guest today is Ralph Burns. Ralph is the Founder and CEO of Tier 11, a leading digital marketing agency that manages over $200 million in annual ad spend. Ralph is also the co-host of Perpetual Traffic, a digital marketing podcast with hundreds of thousands of monthly listeners.

During our conversation, Ralph describes the evolution of the industry and some of the challenges he’s overcome whilst building his business and growing his team. We discuss the rapid adoption of AI tools within the digital marketing space and Ralph shares some insight into the increasing value of the human experience in a world that will soon be dominated by AI-generated content. The world of digital marketing is at the forefront of GenAI and I couldn’t think of a better person to shine a light on it than Ralph. So, please enjoy my conversation with Ralph Burns.

How the Digital Marketing Industry Works

[00:01: 24] Sam: Ralph, thanks for agreeing to share some of your digital marketing expertise with listeners and talk about how you see AI impacting the space. Perhaps just another acronym to add to the soup that already exists: ROAS, CMS, CTA, SEO. Frankly, it's a bit of a black box, but I was hoping maybe you could start us off by giving a quick breakdown of what a digital marketing agency does and how Tier 11 helps its clients.

[00:01:52] Ralph: Yeah, I know. It's great to be on the Not a Threat podcast. This is awesome. It's a great subject and something that is near and dear to my heart. Obviously, the AI component is a huge part of what we do as a digital marketing agency. But I look at digital marketing really simply. It's the easiest way and the most measurable way in which to grow a business. And we just so happen to do that through paid traffic, through digital means, primarily two of the largest advertising channels on the planet, which is Meta, which is Facebook, Instagram, you name it. And then there's Google. And then we also do all the other channels, which is Snapchat, TikTok, LinkedIn, you name it.

But basically, all those are paid traffic channels that just drive traffic, get eyeballs. And in the case of Meta and Google, you've got pretty much the internet covered between the two of those. So, if you just focus on the 80/20 of digital marketing, which is what we try to do, customers hire us to help them scale and grow. The purpose of Tier 11 is to help purpose driven businesses achieve their vision, and you do that typically by exerting more and more influence and getting more and more customers, new customers, and then enhancing their lifetime value. And we just so happen to do it through digital means.

And we just see that it's much harder to do it through means like TV and newspaper advertising - if that's even the thing - or flyers in the mail, direct mail because it's harder to measure. Everything digitally, you can measure and even though the measurement has been diminished, I would say in the last three years due to privacy concerns, it's still the best way to grow a business, in my opinion, there's no question about it.

I started off in sales and sales management in the corporate world and that was basically just helping businesses sell stuff. In my case, I was just a sales manager. And digital marketing is a natural extension of that. It's just a wider influence. When you've got a platform like Meta that covers two thirds of the world's population - which still blows my mind - two thirds of the world's population is on a Meta platform and then everyone uses Google. Google is the internet. If you can harness the power of those two networks and know what you're doing, with a bunch of talented professionals like we have at Tier 11, you can scale and grow any business on the planet.

That's basically what digital marketing is without getting into much of the acronyms, but it's just a means in which to do the thing that advertisers and marketers have been doing for years and years. We're just able to do it with more laser like precision and a hell of a lot of measurement and all those acronyms that you mentioned before - Return on Ad Spend and CPA, cost per acquisition - it's just very measurable. And at the heart of it is a tremendous amount of data and you need a data solution that can aggregate all of that data and give you insights into making good decisions on where you should spend your money and which platform, which campaign, which ad, all of that sort of thing, ultimately to achieve the goal, which is to acquire new customers and enhance their lifetime value.

So that's really what digital marketing is. It's simple to me. The actual intricacies of it is a little bit more complex, but that at its base is what it does. And that's what we do for clients.

The Challenges of Building a Digital Marketing Agency

[00:05:24] Sam: Just before we hit record, we were talking about some of the challenges of building a business. Could you go into a bit more detail about what that experience has been like for you?

[00:05:36] Ralph: Yeah, I think the funny thing about having a marketing agency is that anyone can do it. If you have a laptop and an internet connection, you can say, “Oh, I'm a marketer, I'm a consultant.” And if you have a specialized skill, you learn how to do Google ads, you learn how to do Facebook ads or whatever it happens to be the barrier to entry for a marketing agency or a marketing consultant is super low.

And I think it's even easier today than it ever has been. That's why there's at last count 143,000 agencies on the planet, it's a crazy number. There's 88,000 in the U.S. alone. So, we are not alone doing what we're doing, but I would say the biggest challenge is to actually get good at it because anybody can call themselves a consultant, anybody can call themselves a marketer and then getting good at it and then having the ability to scale it, especially the way that we've done it. We've always been a virtual organization, meaning we've always been remote. Remote now is cool since the pandemic, but I started this thing, in the back of an RV after I was fired the second time in the corporate world and just had an idea.

I was an affiliate marketer. I was basically selling other people's stuff for a commission, digitally, and I said why don't I do this for actual businesses? And that's when the idea was hatched and we ended up focusing really on one platform at that point in time, which was Facebook advertising, which was brand new in 2010, really 2009, 2010, is what it came to be.

We jumped on that bandwagon. It was just me and a VA, a virtual assistant, who I still actually have on staff, believe it or not, and the challenge has been - getting back to your original question - is to scale that from a guy in his basement with a VA to then three people getting my first hire and then 10 people and then 30 people and then now upwards of 100 people and the systems and how you actually go about doing the day to day has changed dramatically from the days of just starting just on my own, me doing it all. And we were talking about this before we hit record - I don't do any of the stuff anymore. I'm literally the figurehead, which is crazy.

But that's been a huge challenge is letting go of all of that and allowing the best people on the planet. We have very solid hiring procedure and way in which we bring people in and indoctrinate our employees to do their best work. And that's really my goal is my job is to allow that to happen and to put people in a position to succeed.

That's probably been the hardest thing because I've never done this before. I've never built a company that's, upwards of eight figures, I didn't know what the hell I was doing when I was first starting. So, figuring that out all along the way, as many management and leadership and business books as you read, the only way to figure out that and to pay that bill for that tuition, is to do it.

And that's the biggest challenge. And usually, the biggest obstacle is myself, self-limiting beliefs, quite honestly. And then hopefully then hiring the best people that do the best type of work and augment the things that I do and do it a lot better than I do it, which is the way that we've built the organization.

The Value of Human Experience in an AI Future

[00:09:07] Sam: I was listening to your podcast, Perpetual Traffic, specifically the recent interview you did with Perry Marshall. You guys were discussing how AI is going to change the world – and also what's not going to change. It was a really fascinating conversation. I'm wondering how that discussion influenced your view of AI.

[00:09:29] Ralph: Yeah, I think Perry is - I think I told him this on the podcast - he's a futurist. I think he's been very dead on with a lot of his predictions. He predicted, when I first met him, that Facebook advertising was going to be the biggest thing ever and it was like nobody believed him at that point in time.

We based our whole company around it back when I first started to know him and we started to work together but the point is that I think he's pretty accurate about what will change and then what won't change. I think the human experience is still going to be there no matter what, and I think one of the analogies that we talked about is that music, for example, is just everywhere.

It's now a commodity I don't even know what I pay for Spotify. It's such an insignificant amount, but I can get any song anytime, anywhere, on any device, listen to whatever I want, and that algorithm and Spotify is an amazing algorithm. It uses AI, it uses data, it knows what I like, it predicts, what I wanna listen to. It's unbelievable. So, there's an example of AI just making my life better. But what is missing is it’s missing the human element.

And the human element is still I crave live shows, and I can't ever see a day when an AI robot is going to perform. I'm not a huge Taylor Swift fan, but I would love to go see Taylor Swift live just because it's an experience. You're sharing that with all these other individuals. And I'm more of a hard rock kind of guy, Metallica is coming to Fenway this summer, I'm going to both shows. I will pay anything to go see them because it's like I listened to it, but it's the experience of the human part of it, the connection with the crowd and the connection with the actual live music that can't be AI-ified.

Although there are bands that are coming out that are completely AI generated; I know some that are in Japan that actually have paying customers, but I have a connection with those artists and that's never going to change. I think that was one of the things that was positive and uplifting about the conversation with Perry was that sort of thing. We still all crave human connection.

AI is just a tool really at the end of the day. Right now, it's basically just a bag of tools. What it's going to be, I think is going to be pretty amazing, but right now it's basically just a bunch of tools that you can use in order to enhance your productivity, but the AI part that's never going to change is the human element.

Sam Altman wrote about the death of the marketing agency because it's all going to be AI-ified. I disagree. He's the smartest guy in AI, but at the center of it all - I see it every single day - you have to have a human brain that understands it all that aggregates it all. And. Yeah, there's going to be elements of it that are going to be enhanced and made better. And probably some people are going to lose their jobs over time, but at the core of it, humans are still going to be at the center of everything, whether it's music, whether it's digital marketing, whether it's your industry that never goes away.

I think that was one of the positive things from Perry and I agree with his belief on it. It's like the human experience is never going to go away, even though we're going to have all these AI things that are surrounding us.

The Current State of AI in Digital Marketing

[00:13:02] Sam: How AI tools currently being used within the digital marketing space and where do you see it going from here?

[00:13:09] Ralph: I see right now; I look at it in three phases. Right now we're in the tools phase. We just have a bunch of tools for us to create content. We create a ton of content every single day. We use 11 tools, I believe. At the center of it, the human directs all those tools.

So, that's really where we’re right now. All these tools kind of work, but none of them work together to create the end product - yet. But all the tools are enhancers. I have one guy that does, we produce a hundred videos a week. One guy does that. He needs the raw materials and then basically he does that and he puts it out on all the different social platforms. Just one example, like our marketing department is using AI, our media buying, and our creative department use the same tools, basically. But we're at that point where the human is at the center of it, directing all the tools. In a few years’ time, there's going to be this midway where there will be a tool that directs the tools. In my opinion, and then ultimately there will be agents.

Remember the little Clippy thing on Microsoft? Bill Gates wrote about this, I totally agree with his view. There's going to be a time, I don't know when that time is - I keep saying it's going to be 3-5 years, but it might be sooner - where instead of 11 tools that we use to produce our content, we use one tool that directs all the other 11 tools and then produces the output. And all the human is doing is basically just saying, “this is what I want.” That is not that far away. And right now, it's a bag of doorknobs for what we're doing, but they're cool doorknobs. some of them come in and out.

We were talking about Descript earlier. It's like, how cool of a tool is that? we can talk about that, how that's just transformed all the things that you do, say digital marketing. All the things you do digitally, like video itself, like writing blog posts or doing any of the things that we do as a digital marketing agency. That's just one tool of like a hundred tools and they're more coming out every single day.

But once again, it's a tool that has to be directed by a human. When the agent is the one that actually does it all, that's when I think you're getting some real leverage for AI. It's like you tell Clippy what you want, and they go do everything for you and then bring it back to you and say, “how's this?” and it's aggregated everything together, or maybe at some point in the future, that agent doesn't even need the human, all the human needs is just to be able to say, “this is what I want.”

It's like, “post 13 videos on different social channels, all with the best clips from podcast 273 and vary it and make it animated with X, Y, Z, like in, subtitles, non-subtitles, make sure that it's in the right formats.” Literally you press a button and boom, it's done. So that will then take out a lot of the human element but I see that day coming. You're still going to need the human at the center of it all, in my opinion, to be able to troubleshoot and to be able to check. But, how much AI actually takes control of that really is the mystery. But I think, with the onset of quantum computing, a lot of the things that are even beyond what we have right now. The power that's with the data that's available. It's pretty amazing, what you can see AI doing in the future. We're not there yet. We're just scratching the surface right now.

And people are threatened by chat GPT: “oh my God, that's going to take my job as a copywriter.” Like that's not even it. That's just the leading edge. That's the thing that Sam Altman wanted us to all see, it's the stuff that's beneath the surface that you don't really know what that is. I think the agents are the things that are really going to be the transformers in years to come. I just don't know what that timeframe is.

Leveraging AI for Digital Marketing Insights

[00:17:26] Sam: Outside of the generative AI tools, are there any other solutions for analyzing the wealth of data - that you're presumably collecting or working with in digital marketing - and providing insights to you and your team?

[00:17:46] Ralph: We're working on one with a partner of ours that will basically give prompts to take action on specific advertising.

For example, there's AI behind every advertising platform. There's basically a Large Language Model at work inside Meta; it's called the algorithm. Google has 72 million data points on every human on the planet, which is a scary thought, but it's true. That’s AI as far as I'm concerned. They're using that in order to target, specifically, people who are in market for certain products. With the prompt being that the human actually does all the work and puts in the prompts and does all the research in order to target that individual. Like all that targeting and all that information that these platforms now have, it's basically all AI.

And Meta and Google are very much, “Oh, wait a second, we've had AI forever” and it's true, they actually have, they just never really called it AI. So, it's basically just a bunch of data that a computer then figures out who's the best person to target at which specific time with which specific ad. So, we've leveraged that for years and years, and that's one of the things that's so great about digital marketing is that option, is that you have so much data in these advertising platforms that you can leverage to get results for your clients.

A lot of people think that's scary and there's been privacy concerns about that sort of stuff that's out there. Having said that, we're leveraging those AI tools within the platforms themselves, 100%. So, the job of media buyer and buying media and digital advertising is easier today than it was probably 10 years ago, even though it's more challenging because there's more people on those platforms.

So, you still need that human to be able to discern what's right and what's wrong. What I think is coming - and we're seeing this right now with a partner of ours which we're an investor in, we’re the second largest investor in this company - they actually are an attribution tool, which basically means they make sense of all the data that's coming from your advertising and give you hints as to which ways you can optimize your ads for the best return on ad spend. They are working on - and I've seen the beta of this - is in essence, AI driven prompts that when you log in, it'll tell you, “hey, campaign 3 seems to be getting more new customers through the SMS retargeting that you're doing on campaign C, you might want to up the budget.”

That is a huge leap forward for us. That's the next phase. That's a tool being utilized to give insights that a media buyer the team that we have that might take them an hour to figure that out, poring over the data and trying to analyze things. But those types of prompts, that's the next step in our evolution. This one company that's doing it is on the leading edge, that's why we're an investor in that company. We use that company's software in order to guide a lot of our day-to-day decisions. So that's the next thing and that's exciting to me.

So, the next level of that would be, there's no human involved. The change would then be made to the campaign or to campaign A or campaign C. Without the interaction of the human, because that's not quite there yet. So that's what I mean. The human is at the center of it all. AI can give you all these prompts, but then you can look at campaign A - actually maybe there's some previous customers that are in that campaign. Maybe that's what they're doing, they're just buying again, as opposed to actually a new customer. So that media buyer needs to be able to look and see and have insights and then give the tool itself that feedback loop to make it better. So that's where it's going from a traffic and a marketing standpoint, it's the AI tools informing the creative person or the media buyer or the email writer, “Hey, your email here got this many views and this many clicks, and you might want to consider using this type of subject line on your next email.”

We have that, but we have to find it now. We have to go digging for it, but having a tool where it tells us that, and then the human is at the center of it and then makes a decision. That's the next phase of things. That's coming, that's within the next three to six months, I would say.

The Unanimous Adoption of AI Within the Marketing Industry

[00:22:37] Sam: How do your employees see AI? What's the general feeling within the business and how do you approach discussions around the implementation of AI with your team?

[00:22:50] Ralph: It's not even a discussion of if you're going to use it, it's just a question of how you're going to use it.

I employ a bunch of millennials. They grew up on devices, you know what I mean? This is just a natural evolution. They don't even think of this as change. So having said that, we had a very expensive copywriter - who's a great guy, still a great friend - he resisted and he's probably 10 or 15 years older than I am, and we just found that wasn't in line with our ethos as an organization and we parted ways. And he lost his job for AI because he didn't adopt AI.

I think I'm in a little bit of a bubble in the digital marketing space is that everyone uses it. No one resisted it. You know what I mean? And even though we did a complete transformation of our staffing in 2023, nobody lost their job over AI except that one copywriter, literally, because he wouldn't do it and he was just old school. So, that's not because he was older, but because he just didn't want to adapt. He said, “this is just not right, I disagree with it.” I'm like, “sorry, this is the way that it's going.”

But we have an AI channel in our Slack for all the tools. It's hard to keep up because everyone's so excited about it. Nobody sees it as a threat at all. Maybe they do and they're not telling me because, as the boss, everyone gives you the soft soap, suck up. But it's just a part of what we do really. I'm a little bit skewed in our industry, I would say.

The Current Limitations of AI-Generated Content

[00:24:38] Sam: Have you noticed a difference between human generated content? And AI generated content in terms of the results that you're getting for clients?

[00:24:48] Ralph: If you look at AI generated content specifically, you can't use it straight out of the box because anything that you put into Bard or ChatGPT is taking the good and the bad from everything on the internet if you really think about it. That large language model is extremely large. So, I would say ChatGPT can do maybe 80 percent of the work for you. We have a huge part of what we do as a research side of it. When we bring on a new client, we need to figure out, “okay, what messaging - old school advertising - what messaging, what hook is going to get them to pay attention and stop the scroll?”

If you think about what we do. If you're scrolling through Instagram and you see a sponsored ad that's probably us. What is going to do this? Stop the scroll. That unto itself is an art and a science but it starts with figuring out who is your avatar, what their pain points are, what their desires are and really understanding that and then crafting a message around that. And that comes from a lot of research, like going deep.

The AI tools, like ChatGPT do that extremely well. So, they can do a lot of that heavy lifting for you because they're aggregating bazillions of data points. And if you put in the right prompt, you can get, “Hey, I've got this, supplement that increases testosterone, for men ages 35 to 45 the name of it is X list all the competitors tell me everything about all the avatars and all the types of individuals. I need seven different individuals or avatars that would potentially buy this product, what their names are, where they live, what their marital status is…” these are all prompts that you would prompt into ChatGPT.

So, we use this on a day to day basis to do a lot of our research but a lot of that stuff will come back and it'll be total crap. So then it's a matter of the human doing the work and then refining it. I got a memo from a university a day or two ago, and I knew it was written by ChatGPT because it was just total crap. Like they just did it and then sent it. So, if we're doing that as an ad agency, I need to be firing some people, seriously, because you still need an editing process. So that's if you're using ChatGPT, because ChatGPT pulls from the entire World Wide Web - the good, the bad, the not so good, usually the crappy - and a lot of it isn’t actually true so you've got to be careful there, too.

So, I think it's a good way of starting. Once again, it's a tool to start. The blank page syndrome if you're writing content literally doesn't exist anymore. If you have ChatGPT or you use Bard - I use Bard probably more than ChatGPT - so there is that.

The Powerful AI in Meta’s Ad Platform

[00:27:54] Ralph: Having said that, if you take that Large Language Model and squeeze it into an ad platform… the AI generative tool for ad copy inside Meta right now is un freaking believable. Sam, it is absolutely incredible because what it'll do is it will be pulling data from just the Meta platform and the page and your best ads and what your competitors are doing. So, it's a smart, Large Language Model. I call it a midsize language model, really, because it's pulling from its own ecosystem. That stuff you would never be able to tell. I could, all right, “here's my ad copy that is killing it right now on ad number one, create” - there's literally a tool inside the Facebook ad platform Meta ad platform says, “create three other headlines that are similar to this” because Meta knows I'm trying to get a website conversion, I'm trying to get a lead, they'll pull from what's worked for your page, what's worked for you, what's worked for your competitors. It's unbelievable.

So that kind of stuff is helping us just inside the platform itself, or it has a generative image tool. Now, you can use DALL-E or you can use any, there's a ton of tools that are out there right now to create images, but the image generator inside the Meta ad platform - it already knows what's converting and what does well. So, when you use that, or it fills in an image or does a background, it has an understanding of what's converted before to make it even better. These are the tools that make our job a hell of a lot easier.

So, I don't think anybody's going to be losing their job over it, but I see those tools as multipliers, so I don't have to hire maybe as many media buyers because those media buyers are just all of a sudden going to become assassins. They are going to be so effective at their role and the ones that adopt that and figure out ways to leverage it are the ones that are really gonna ultimately succeed.

I have a friend that created a tool - he's an incredible copywriter, he was on our podcast just recently - and he basically just created his own Large Language Model with all the most successful copy that he's ever written. The most successful video sales letter scripts he's ever written and distilled it down into a tool that creates ad copy and sales letter copywriting and video sales letters - or VSLs is what they're called - video scripts. And it's unbelievable, but it's taken from his best stuff. If you're putting in the best, the cream of the crop into your Large Language Model, all of a sudden it's even more powerful.

And I see tools like that. It's that thing, when I saw it, I was like, “Holy crap, that's a whole next level” and it was written amazingly well and then you could have 17 different ways in which to say it. “Oh, I want to sound more esoteric. I want to sound more to the point. I want to sound like a fifth grade, writing style,” like all these additional prompts and it changed it like that and it was so well written. We did a demo of this on the podcast. My co-host and I were astounded at how great it was. Like, ChatGPT can't do that. It could maybe if you gave it enough prompts and it started to really learn and understand, but nothing like that. So, it's really garbage in garbage out, gold in gold out. If you're putting in the good stuff, the example that I said about the AI generated headlines for Meta or this guy's tool for copywriting, that's when AI becomes really fun.

Getting Started With AI in Your Business

[00:31:54] Sam: If you were giving some advice to somebody who's just looking to start deploying AI within their business, how would you recommend they go about that? Where do they start?

[00:32:05] Ralph: I think they have to think about what the end goal is really, what are they trying to achieve, specifically? Is it a CEO trying to write an email to his team, or is it a salesperson looking to write an outbound LinkedIn message to a potential prospect, or is it a school teacher trying to figure out what her curriculum should be for the coming year, for the next month? You have to look at what your goal is first and then figure out what tool might actually help you. There's probably a tool out there that would help in all those cases and that's really a Google search away.

Figure out the thing that you don't really like to do all that much or that takes up the most of your time. Just do a Google search and say, “AI tool blank” and chances are, there's probably something that can help you and it can probably shorten your learning curve and also get you acquainted with the tool that actually helps you from a productivity standpoint.

Obviously, there's ChatGPT and there's Bard and all those tools and I think ChatGPT, the better version is only 20 bucks a month. I think Google Bard is free as far as I know.

Going back to our original conversation, I always think to myself, “there is somebody out there in the world that does all the things that I hate to do”, and chances are, there's probably an AI tool that helps you do the things that you hate to do. It can probably do it better, or at least can get you an 80/20 so that you don't have to spend 3 hours doing it, you might be able to spend a half hour doing it instead and just go into it that way. Whatever that urgent pain is, there's probably a tool out there now.

Not all of them will be around in a year or so. Even some of the ones that we've used in the past six months I'm like, “where did that one go?” you know because they just lose funding or they can't figure out how to monetize it, which is fine, but I would say that this is probably the way to go.

I would say everybody writes emails. I actually made it an actual policy inside Tier 11 that you have to use Grammarly. I think it's the best freaking tool that's out there. It's a total AI tool, but it's a total AI hack and you can write emails, letters, it just makes everything better. It just takes what you're writing and then it just makes it even smoother and better and fixes your grammar mistakes and all that. That's probably the best tool I would say for somebody who's a non-AI person. Just try that out. Get the $5 a month thing, it's an amazing tool because everybody writes - unless you're a common laborer or whatever, but they're probably not listening to this podcast here Sam is my guess, but maybe they are!

The point is, that's one tool, you'll be like, “holy crap, AI is cool because it's making my life better, and it's saving me time, and it's making me sound smarter, more eloquent, or, well spoken”, and hey, words matter and we still are society that runs on the written word. There's obviously voice and video and everything else, but I would say that tool right there is the one that you should start with.

[00:35:24] Sam: Awesome. Ralph, thank you so much for the advice and really appreciate you coming on today. It was great to speak with you.

[00:35:30] Ralph: Appreciate you having me on, Sam. It's awesome.