Here’s a video/podcast we recorded recently on the topic of AI for B2B marketing and sales. This is part of my work with Gerson & Associates, a company for which I’m the Chief Growth Officer.
Brian: [00:00:00] Welcome to another episode of Executive Edge today. We’re going to talk about the future of AI and B2B marketing and sales. And everybody seems to be excited about AI right now. It’s a. Big topic of conversation. There’s a lot of things coming out every day. It’s changing and there’s something new Chris was just telling us before the podcast about Hey Gen which will have like a video avatar.
I might be one right now. You can’t tell
Chris: yeah That’s the idea is that that could that could be the case? You upload either an image of yourself or a video of yourself. You record your voice. It has prompts and what to say. And then you have your avatar created. And then you can type whatever you want yourself to say.
And a video of you will pop up. It takes roughly an hour to set up.
Brian: Yeah, I’m sure that’ll get better. Like [00:01:00] everything else with AI, who knows in two months, it’ll be like 15 minutes.
Chris: And don’t, don’t forget Synthesia and Canva already have the avatar AI’s, but it’s not, those aren’t avatar AI’s of yourself.
Those are avatar AI’s where you select. Who or what you want the avatar to be to present to the individual. And Canva’s avatar AI, I have not played with it yet, but supposedly it’s pretty good.
Randy: Did they have a Fred Flintstone? I’d love to use that one.
Brian: Oh, you wanna be Fred Flintstone?
Randy: Old reference. Right.
For, for a lot of the younger ones.
I here to look up.
That’s an old reference.
Chris: I have talked to somebody that’s building an, an agent, you know, and a ChatGPT agent and they’re doing it in the form of Mr. T.
Brian: Okay. . Well, and for, for Gen Z, let’s just say maybe it’s Harry Styles or whatever, but
Chris: Sure, sure.
Randy: Harry would be good.
Brian: And this is, I mean, this is fun stuff, but it has real world applications. So especially with what [00:02:00] we’re already talking about scaling, whether you want to scale to get more messaging out to more contacts you want to be able to keep that personal. And again, you’re using your face. So it looks like you That’s a lot of work.
How many how much time would it take you to create? A hundred videos of yourself saying things just most people salespeople don’t have the time to do that.
Brian: So this is a really cool way that marketers can help salespeople. What else so like predictive lead scoring potentially? A lot of those, when it comes to a lot of things we want to talk about here, hyper personalized content, AI powered account selection, AI generated content.
talk about the hyper personalized content. Yeah. So we were just talking about ABM marketing, accounts based marketing. And there are now, I think over a dozen or more hyper personalized ABI ABM AI personalized content companies. [00:03:00] What they do is they evaluate and watch when somebody Reaches your website.
Once somebody reaches your website, they know what company it’s coming from, right? Using reverse IP lookup. Once they know the company, let’s say somebody from Lockheed Martin comes to your website. It says, I know somebody from Lockheed Martin. It’ll check your HubSpot database. Immediately on the fly and says, is there somebody from Lockheed Martin in our CRM?
Oh, yes, there is. By the way, that person’s in the middle of the funnel. I’m now going to show them website content based on the fact that they’re in the middle of the funnel. I already know that they’re aware of the web, aware of us, know who we are, blah, blah, blah. I’m going to skip to middle of funnel content.
Or if they’re at the bottom of the funnel, it skips to the bottom of the funnel and immediately. pings the sales rep and [00:04:00] sends an email to that individual if they’re not in the funnel, it knows that and will serve content based off of top of funnel, knowing that they’re in the D. O. D. Aerospace industry, and I’m going to show them relevant content as it relates.
So it does all that. Maybe HQ is one of them that does it really well. They’re starting to take off AdRoll. Has a division of the company that is B2B ABM focused. The majority of those products are around 80 to 120, 000 a year, only available for yearly contracts. The reason for that is that they claim that you cannot test it for one or two to three months because it takes longer for it for that.
To actually start working and that they think if you cut it less than [00:05:00] six months, less than a year, you’re cutting the testing too early.
Randy: And Chris, that brings us back to a lot of people ask this question, even of us, right, when we contract with them, why does it take, you know, why can’t we just do a three month test or a, you know, two months because you can’t get enough data to know when you’re talking about learning engines, which AI is, it’s a learning engine, right?
It needs data input to learn from and, and learning engines need a lot of data input when it’s that highly personalized. And so that would take a long time. And of course, there’s a lot of programming cost because, you know, once you set an AI off, you just don’t leave it alone. They’re constantly lack of better words, reprogramming or modifying it.
As it goes on, most of these learning engines, most people don’t realize that they just don’t learn completely on their own, there’s still human involvement. The other thing to mention of what you just said, Chris, there’s the other approach, and I haven’t seen any companies put it all together yet, but I’m sure there’s one out [00:06:00] there.
Just haven’t run across it yet is there’s the IP reverse append, and then there’s big data. And as you remember, we had a way, although no one’s sharing who it is yet, but there’s a way on your website, somebody comes to it, and through aggregation and compiling of big data, meaning Facebook, Google, online behavior of an individual, like let’s say Brian comes to the Gerson and Associates website, it knows it’s Brian, it knows it’s home address, it’s cell phone number, and it’s expensive And then we can send out personalized content.
You talked about top of the funnel. Brian just visited the website. He’s not in our CRM database at Lockheed Martin, but we know he works for Lockheed Martin, and we know how to reach him through social media channels and other places again. That’s it’s expensive to do that, and I haven’t seen I’m assuming that knowing how things work that Facebook, Google and the big boys are working on that solution as well for ABM because they already have the way to find out who that is through their big data.
So that’s just another way that [00:07:00] some of this is and the thing I love about this is that always comes down market to the smaller companies at the way this is accelerating, maybe only a year later where it used to be five.
Brian: already, I mean, there’s so much possibility, but there’s already ways, you know, there’s already that little, maybe you don’t have a Swiss army knife, but you got seven different tools out there already that you can use for these things. And specifically, we’ve, we’ve already talked a little bit about content, but many, many B2B companies have a content problem and we’re not even talking about.
Getting them to level where they’re personalized to each account or scaling what we’re talking about. Do you have the basic, like Chris was saying, do we have the mid funnel content? What is that? Is it a PDF? Is it, do we have multiple things? And AI has so much promise for, and it can already do so much.
To help create that content your your b2b marketers can use this stuff to be more productive than they ever have been before[00:08:00] or you can have us help you with that, but that needs to happen because more and more people the competition level is going to go up here as the capabilities go up
Randy: Yeah, and Brian to tag onto what you said no matter how automated that gets you still got to have a human eye on it And as we’ve noticed, we see these companies just, I’m going to have an automated content Chris and I get this when we teach at the college level of the MBA marketers and they don’t even edit it and it has misspellings using the wrong word in the wrong context wrong picture, but thinks it’s the right picture for the content with the wrong headline, didn’t match those up properly because it’s AI, it’s learning.
They got to always remember this is it’s a learning engine and it takes a human still to. checkbox. That one’s good. Let’s send that one out the door. So let’s not forget the basics of you can’t just set this. It’s not a set and forget.
Brian: And for example, you can take something like chat, GPTs, new GPTs thing, which is poorly named because it’s confusing with chat, GPT versus what is a great job, but [00:09:00] GPTs allow you to not only have custom instructions for something, but also upload content so that, you know, because the, the, the AI knows.
What it’s been trained on, but it doesn’t know your niche. It doesn’t know your customers and their details. And of course you got to be careful with what you’re sharing and what you’re not sharing that’s proprietary. But if you really want to customize this stuff for your specific usage, there are ways to do that out there.
And it’s just, it’s a constantly changing field. It’s changing and developing faster than anything I’ve ever seen. And, and that means there’s new obstacles and some of them are hundreds of tools, which ones should I use? I don’t even know what they all are. Or chat GPT right now being too popular.
It’s not working right. Because it’s their, their network is being crushed by all the interest. So you need help. You need someone to help you with this stuff.
Randy: Yeah, and Brian, I think that you brought up a great point about the help. You can’t just look at AI as a cost cutting [00:10:00] measure. Because if you look at it for just that you will you’ll crank out more stuff quicker But I always say again, I’ll use that technical term.
I like more crap Will go out the door and you might again tarnish your brand or offend people Right without really editing some of the stuff that’s going out. We’ve seen it from big box stores do that already
Chris: You’re, you’re right in a big way, Randy, I believe. I believe the first step in the LLM, the chats and the bards, is just to get used to using it and get doing things faster.
But if done improperly, what you’re saying, Randy, is you could see a decrease in response. Right. So if you do it right by adding right now, we have to add the human influence in it and make sure that it’s optimized for conversions. Then we’re going to start getting a higher response, getting the ROI we need.
It’s not all just about saving time and money. It’s about getting [00:11:00] More sales and there’s
Randy: Chris on that. I mean, you’re so spot on. I’m going to give an example. I saw on the news yesterday. That wasn’t AI generated. That just shows a campaign was generated ahead of time and then times changed. It’s a big box store.
So I think it was either Walmart. Our target, I forget it was big grocery store, had a ham sale for Thanksgiving and it came in a bag that said, happy Hamas, ham dash mass like Christmas. Okay, exactly. So for today’s times, they scrapped the campaign. Now, if an AI generated that and a human wasn’t looking at it, so the human caught that mistake and they pulled it off the shelf.
But again, I’ve think about the speed of AI creating stuff in real time and real events going on worldwide. You could have a clash again of information that is Diminishes response and on top of that diminishes people wanting to talk to you because your brand value goes way down. If you have any brand value, you’re not a big company at all to start with.
So you can be very, very [00:12:00] careful. Again, human eye has to look at this stuff. I agree with you. Get used to using the technology, but don’t forget the human element. Well,
Brian: I used to say, we used to, we were calling it AI shepherding, and now I want to call it wrangling because it’s like a fricking alligator just thrashing around on you.
And I think, I think we’re getting into like, what I would call, like, there’s all these people on there, on Reddit, talking about the singularity and when AI will become intelligent and stuff. And they, most of them don’t have a real understanding of like, What business is really like the complexities and the human element of it.
So we’re getting into like the messy period where things are still growing, developing. They’re not all finished. You know, people are, are complaining that, Oh, I was basing some of my, my work and my business on chat GPT, and it’s not working now. So we have to make sure we have reliable solutions and we’re picking those over others.
But then there’s two different levels of of AI here. We’ve got [00:13:00] what an enterprise can create with programmers, creating custom stuff, right? Whether it’s with open AI or not. And then you’ve got what, you know, a small business might use on a a B2C basis, like these AIs that are being created for, for smaller businesses.
Randy: Yeah, that’s a really, really good point, Brian, for that. And, and again, not to use a cliche, but you know, the big wave of AI has hit everybody. It is the hot buzzword, right? At the moment, at the moment, right? And, go ahead, Chris.
Chris: I’m sorry to interrupt. By the way, I was just going to say, I I normally don’t play around with Google ads.
There are other people that are much better at it than I. But yesterday, I decided, you know what? I wonder, I wonder what the platform looks like now. I haven’t logged in for a year and a half. You know, I know they added some stuff. Well, it’s a completely different system. And it says, who are you? [00:14:00] You type in your URL.
And then it pops up. Would you like us to create the ads for you? Yes or no? Okay, I got to see it. Yes. Here are 17 variations. Boom. These are the keywords. Boom. It was five minutes. And it was. 85 percent maybe even 90 percent there. I’m like, holy cow, it is advanced significantly in the last six months.
Brian: I will say, yes.
But as a, as a guy with another agency that has Google ad services, when you get to the complicated levels, like sometimes AI is deciding the wrong thing or automatically doing things that, that backfire, but yeah, they are way better. Like I remember 10 years ago, they were trying to create, they were trying to do this back then and it would make all these errors, but yeah,
Chris: the other thing [00:15:00] that interesting is Google and Facebook are not allowing us.
As markers to make some decisions anymore. True. That’s interesting. They’re like, sorry, we’re going to decide how to serve these ads. We’re not going to let you choose. And I’m like, holy cow.
Brian: Sometimes that makes sense because they understand how their engine works. But other times it’s because they don’t want to make money.
And I mean, historically, these ad platforms have always been like, let’s get you more clicks. I’m like. I want more sales. They’re like, we can get you
more clicks though.
Randy: Which is what happens. We all know that from talking to the ad managers at different platforms. I won’t pick one out specifically.
Chris knows what I’m talking about. We get on the phone and they’ll tell you, well, if you spent more money here, you know, then you do and you don’t get any more sales, right? Yeah. It only improves, if you’re lucky, one part of the funnel. And so I can’t imagine, Brian, and you hit the nail on the head, that they’re AIs.
on and actually, Brian, you remember this? We were beta for the AI for the original one for Facebook and the very, very, [00:16:00] very beginning with our client. Right. When you and I first met and we both were like, what happened to our money? It’s literally went through tens of thousands of dollars like overnight and we got nothing for it and then they had to apologize.
So I just want everybody to be. It’s awesome. I think this stuff is awesome. Of course. We love tech, right? It’s not terminator stuff yet. Like everybody thinks in their mind from the movies. Don’t think that or some sci fi movie It’s not there yet or might never be there in this capacity So i’m going to push this again and brian you said it got to have a human in there like chris said 85 maybe 90 was there It just tells you, you still need a human overlooking it, even though…
Chris: Still needed. Absolutely.
Randy: These platforms are taking some decisions away, you have to make the right decisions that you can make to get the most bang for your buck.
Chris: And there were some ads that were completely off base. Yeah. You know? Like Hammas? So you
Randy: have to… Ham mash Yes,
Brian: exactly. Not quite a blind swirl, it’s more like a…
I don’t [00:17:00] know, half blind squirrel.
Randy: Squirrel with one eye, just only has one eye. Can’t see the other perception. But it’s fun, good stuff, you know, and we like to test this stuff. Cause you just, you never know on the next test that you do. And that goes back to everything else. You always gotta be testing.
Using these technologies to test to see what they can, and more importantly, what they can’t do. You know, so you don’t over promise. It’s the last thing marketers got to be aware.
Brian: Last thing I’d say is it’s a lot like the beginning of social media where a lot of people were like either completely unaware it existed or just couldn’t see it, didn’t see how it would fit and so on.
And then becomes eventually just unassailable. It just becomes a foundational part of what you’re doing. And everything keeps changing. So
Randy: I think you’re right. I used to think social media was, I grabbed the, the LA times that I got walked over to my neighbor and we discussed the paper that was in social media.
Brian: Well, I’m probably with more social than [00:18:00] what we have now, but yeah,
Brian: well, that’s that’s a little bit of the future in AI for B2B marketing and sales. And it’s another episode of executive edge. Thank you guys for listening and please tune in for the next one.