[00:00:00:01 - 00:04:07:01] Daniel Burstein welcome again. Transparent Marketing part two, how to earn the trust of a skeptical consumer machine. You are in an AI guild session. I am going to just jump right in to get your brain working right away and get value from every minute we have today. So let's just start with it. Let's start 15 years ago, right? I'm going to get to the presenters. We're going to slow down once we get there. We're going to speed up at first so we can just kind of pull your brain right into this. Let's start 15 years ago. Before I did this AI garbage, we didn't hear about it, right? What was going on? We were still trying to please the machine. We were trying to please this algorithm, the giant scary Google. And I love this story from 15 years ago. This guy, Marcus Sheridan, you might know him now as the sales lion. We started working with him back in the day and he spoke at some of our conferences and shared some of the stuff he learned. This was really interesting. They just had a little pool company back then, a little pool company in Virginia, and they were able to rank really high in Google at the top of Google. How did they do it? They answered people's questions. Scary questions for us as marketers and entrepreneurs, but questions like, how much does the darn thing cost? So they, blog posts answered these questions. They were killing it in Google, ranking really high, right? Now it was 15 years ago. That was a different algorithm. Maybe it wasn't AI yet, right? But I thought of this when fast forward to today, I did this case study recently on marketing share from maybe a month ago. And now it was a software company. Now it's a SaaS company. They said pretty much the same thing, right? Transparency worked back then 15 years ago with that. They said the same thing. They said, hey, we had this page, our pricing page. It's kind of focused on features. We had a little bit about pricing. But a lot of it was like book demo, if you want to know about the pricing, right? Know what they do. They talked to people. I'm sure you do this too, when they actually had sales calls and stuff, they'd always ask, how'd you find us? One thing they learned, which I'm not sure will not shock you if you're in an AI guild session, is people started going a lot more from just doing that Google search, right? How much does the pool cost? Started doing AI. And they were asking this question. They were saying, in general, right? Everyone's a little different. That's how AI works. But they were asking for employee advocacy platforms for LinkedIn for a team of 20 with affordable pricing, with affordable pricing. So what's that big AI machine brain going to do if you don't have your pricing on your page, right? It's a lot less likely to recommend you. So what you did with all everything we show in these slides today, by the way, we're going to email out the slides afterwards. There's a source link you can see on the bottom. You can read the full case study if you can't see all the details here. But I just want you to know at a high level, what did they do? They basically made their pricing a lot more transparent. They put it out there. They had clear FAQs. They made it clear. They made it transparent. They made it helpful. And what happened, just over six months, they went from getting 96 demos when they were focused on paid ads to 216 demos when they focused on AEO, GEO, right? So two examples from the pre and current AI era, but they taught the same thing, right? And what if you're saying, oh, I get it. I can leave now three minutes in. Thanks, Dan, you saved me time. All I need to do is slot my pricing on my website and I'm good. No, that's not the lesson I'm going for, right? There's a lot more to think about. I'm not going to get in with pricing. There's a whole cost value spectrum of when people are walking through your funnel, when you present costs, when you present value. There's all that stuff. Pricing might be right. It might not be right, but there's a bigger lesson. This is what we're going to spend the next 57 minutes on, right? Humans need transparency and so do machines, right? LLMs just like humans need and reward transparency. And we're going to get into some of the things that Flint taught decades ago and some of the things he's learning now and that Andy's learning now too. And I think it comes down to this big idea to help you figure out how to do this. And this is what a realization I had the other day is AI has made us better marketers. Not because we can just throw a bunch of stuff in AI and it'll throw some headlines out and we can spit it out. No, no, no, not for that reason. AI has made us better marketers because it has forced this kind of transparency onto us. It has forced us to serve this customer because it is, it can be the ultimate skeptical customer acting on the customer's behalf. All right, so this is what we're going to get into today. Let me tell you who we're going to learn from. Joining us from Cleveland, Ohio right now, Andy Crestedina. Andy, I've seen his content for a long time. You know, he's kind of someone you see out of the corner of your eye and you're like, he's doing some pretty [00:04:07:01 - 00:04:16:18] Daniel Burstein there. What's going on? And then I had recently a chance to reach out to him and I'm so glad we did. I got to know him a bit and he has some really good stuff. Andy, you want to tell us where you're joining us from today and how are you doing? [00:04:17:10 - 00:04:53:07] Andy Crestodina I'm Anthony Linden-Ohio. I'm at the Marketing AI Conference, MAE-CON. It's a big event to people just like us who are figuring their way out in this era. And I'm honored to present this. My third year presenting here end up, I give a lot of presentations, like 100 a year if you encounter webinars like this. So yeah, many, many years in digital. I'm the co-founder of an agency focused on lead generation, which includes both traffic and conversion. We do have lots for B2B brands and have known of Clint forever. Everyone who's been doing this this long knows of Flint. But now, so it's an honor to be on this and I'm excited to chat with Flint. [00:04:54:07 - 00:05:08:21] Daniel Burstein It's really awesome to have you here and Andy is the author of Content Chemistry, the Illustrated Handbook for Content Marketing, right? So he's going to talk to us a lot from that content side, that traditional SEO side that's moving into AO, GEO, right, the getting people here side, right? And [00:05:08:21 - 00:05:12:21] Daniel Burstein have, call him our returning champion. If anyone's been on the AI Guild before, you know, [00:05:12:21 - 00:05:18:05] Daniel Burstein with him for 17 years. He's been doing this since the 90s, right? So [00:05:18:05 - 00:05:49:16] Daniel Burstein some of these new AI folks coming out of the woodworks or working out of the garage, who knows what they're doing? They got their level of pages set up, right? He's been, we've talked about, this is his third digital revolution he's been through. He's the author of the Markers, Philosopher, 40 Briefful Selections on the Power of Your Value Proposition. He holds 10 conversion related juristics. One of those, which is your juristic, Andy mentioned in the article that I read from him when he talked about motivation and how it relates to AI. But on general, how are you doing today on top of the mountain? Flint, he's on top of a mountain in Montana. [00:05:51:03 - 00:06:02:13] Flint McGlaughlin Well, I'm good, I'm at my ranch in Montana. I'm at this picture and in the spirit of transparency, I'm wondering if I need to update it. I think I look a bit older in real life than I do in [00:06:02:13 - 00:06:08:17] Flint McGlaughlin picture, at least my beard is white. So maybe I'm already violating the premise of this [00:06:08:17 - 00:06:15:14] Flint McGlaughlin But fortunately, we have Andy here to make up for my age and decrepitness and [00:06:15:14 - 00:06:17:08] Flint McGlaughlin his youth and vigor. And [00:06:17:08 - 00:06:19:10] Flint McGlaughlin so I'm ready, Dan, when you [00:06:20:16 - 00:06:30:14] Daniel Burstein Well, I'll give Flint one out. I'm mostly going to tease him the whole time, but I will give Flint one comment with that. As I mentioned, he's one of, I think the young people call him "Uncles" now. I don't know if you've heard that term, they call him "Uncle." [00:06:30:14 - 00:07:57:17] Daniel Burstein He's one of the old forefathers of digital marketing. I have seen no one with more excitement and vigor about this new AI stuff. It would be easy for him to be a crotchety guy like me and be like, "Eh, AI, whatever." This guy, he is just researching and working on it day and night, and he's so excited about it, and we're going to share what he's been learning. But there's also you. I don't want to overlook that fact. We have the AI guild with us today. Feel free, use the chat, share what you've learned, ask your questions. I've got a whole bunch of questions for these guys. I'm going to be looking at the chat too, though, to see whatever questions you have and to treat it, try to intersperse that through. If you have some, again, share what you've learned. If you can, we see a few smiling faces. Feel free, turn those cameras on, right? We can all present and get to know each other a lot better. We're in the same place at the same time. If we can actually see you, we might be able to talk and interact a little. We'll see what we can do today. But when you have that camera on, then you realize, I'm not just watching some YouTube replay. I'm here with these people. I can interact. I can share. I can get to know other people. So I really encourage that. It's good to see in the faces we have today. All right. With that, let us jump into what we're going to talk about today. And we're going to start again. We're not starting in the AI era. Let's look at this differently. We're going to start in the year 1868 with this really cool poem from Emily Dickinson. I'm not going to tell you why we're talking about it, but I just want to ask Andy about it because when we were emailing about it, Andy got super excited. So Andy, give us a sense. Tell me what you learned from this poem specifically that has helped you in your marketing, communications, advertising career. [00:07:58:17 - 00:08:55:24] Andy Crestodina I have loved this poem for so long. And when I stumbled across this influence content, I got excited about it. And so yeah, it started a conversation. So it's short enough I could read it, I won't. But the idea is that indirect communication is the only effective communication for certain truths. And that's an important lesson for marketers because a lot of marketers are overly direct and overly branded. And the marketing sort of sounds like we're number one or first time every time. And that's especially a problem now because taglines don't train the AI. Also, I think at a bigger, almost philosophical level, this poem is relevant because in the AI era, there's sort of this indirect approach that works best. There's an intermediary in between us and our audience. If your audience is talking to a machine, then the truth has to dazzle gradually. You're going through this kind of indirect method because there's this abstracted layer in between. But yeah, there are many takes on this. It's a profound, [00:08:57:21 - 00:09:07:04] Andy Crestodina brief little chunk of language that sits very deeply with me. That's one perspective on it. Flint, I know that you came at this from some other angles. [00:09:08:05 - 00:09:13:15] Daniel Burstein If there's a specific line in here that really stuck out to you, the truth must dazzle gradually. What was it about that line? [00:09:14:05 - 00:09:16:17] Andy Crestodina Well, it's-- yeah, Flint. [00:09:16:24 - 00:09:18:08] Flint McGlaughlin no, after you, Andy, please. [00:09:19:12 - 00:09:21:05] Andy Crestodina Well, it's that [00:09:22:19 - 00:09:27:08] Andy Crestodina too bright for our infirm delight. You can't simply say that you are-- [00:09:29:13 - 00:09:35:13] Andy Crestodina you have to support your claims. You can't just say you're number one. You can't jump to the conclusion at the beginning. These [00:09:36:23 - 00:09:59:01] Andy Crestodina homepage headers that say best in class, first in class, best industry leading, that's not how to do marketing. Make your case. Build it up. Build your content so that it emulates a sales call and dazzles them gradually through a series of messages, low ambiguity, well supported, so that the person arrives at the conclusion without being hit over the head with [00:10:00:23 - 00:10:08:07] Flint McGlaughlin Excellent point, Andy. I'm a marketer, let me just say this to you. A simple way to think about your task [00:10:09:14 - 00:10:21:15] Flint McGlaughlin and a counterintuitive way to what you might have heard if you got your masters in marketing. Is that the task of the marketer is not to force a claim, it is to stimulate a conclusion. [00:10:22:17 - 00:10:43:09] Flint McGlaughlin Listen to that differently. When you force a claim, the other side of this equation naturally feels that they must defend against this. When you provide the right signals so that people draw the right conclusions, they will defend their own conclusions about you even in front of others. [00:10:44:09 - 00:10:57:11] Flint McGlaughlin So the marketer has a far more subtle task. And candidly, if you look at what has gone wrong in marketing, you can see over and over again that we are forced to review claims that are often deliberately [00:10:59:03 - 00:11:41:03] Flint McGlaughlin obscure because they're not accurate. You say I'm the leading ex when you want to imply that you're number one, but you know that you're not. Or the claims are not strictly honest. And you could get away with that with some human beings, but you will not get away with that with the machine. And that's the significance of what Andy just said. What he said is profound. If you were to stop right now, look at your own web page and say what are the foremost important conclusions I want someone to draw when they're done reading my web page. You could work backwards from the conclusions to the signals and come up with a much stronger presentation. Dan, I'll step aside. Go ahead. [00:11:41:23 - 00:11:50:02] Daniel Burstein Yeah, well Flint, so I think where that really helps is this line and other things we're gonna fast forward now from 1868 to the year 2003. [00:11:51:04 - 00:11:52:03] Daniel Burstein They made their way into transparent marketing. And [00:11:53:08 - 00:11:59:18] Daniel Burstein so, soon we're gonna kinda get into the AI examples and see those, but I wanna make sure that they're based on a good foundation. [00:12:00:18 - 00:12:11:04] Daniel Burstein So give us a sense, you're talking about some of the things in the web page and what inspired you to write transparent marketing back there? And what were you trying to teach or how you're trying to shake marketers back then? [00:12:11:17 - 00:12:19:07] Flint McGlaughlin Well, marketing has changed over the years, and it's become a game where often confusion serves the marketer. [00:12:20:06 - 00:12:24:19] Flint McGlaughlin But if you have a proper value proposition, then clarity trumps persuasion. [00:12:25:20 - 00:12:34:21] Flint McGlaughlin And the point of this article was to introduce something that came right along with it, which was called the prospects protest. And also [00:12:34:21 - 00:12:49:09] Flint McGlaughlin marketers creed. The three together were a battle cry. And in that cry, we were saying you're approaching this wrong. The world doesn't listen to you anymore. And if you don't change your approach, you're not going to get your message heard. [00:12:50:09 - 00:13:01:10] Flint McGlaughlin This article was that rallying cry. And this point that Dan has on the slide. If you tell me what you can't do. I might believe you when you tell me what you [00:13:01:10 - 00:13:02:01] Flint McGlaughlin do. [00:13:03:02 - 00:13:28:22] Flint McGlaughlin Did you know that right now because you're still running our research program inside of meclab's Institute and we've done a bunch of testing on what's become known as in some circles, a EO, how to get the machines to put you in the consideration set and recommend you did you know that one of the key pillars of the algorithm that drives that recommendation is being able to determine from your website who you can't serve well. [00:13:29:24 - 00:13:40:10] Flint McGlaughlin And yet most of us never want to put that information on the page. It just goes against the grain. We want to tell you how amazing our solution is we make lots of claims. But I have [00:13:41:24 - 00:14:17:09] Flint McGlaughlin relationships in my life I have three children, all of them single. And they're all going to make major important life decisions and when they're making those decisions with people on the other side of that choice. They need to understand the second half of this quote about that person about that offer about that choice, for instance, they're going to make with their financial investments. We need to be honest about what we can't do. It might be that people start to believe us when we say the part that we can. [00:14:18:18 - 00:14:25:15] Daniel Burstein And the thing I like about that is some of the things we talked about obviously have changed so radically with AI. But those things that were true that when Flint was writing them in [00:14:25:15 - 00:14:26:03] (Inaudible) [00:14:26:03 - 00:15:07:19] Daniel Burstein when you look at what AI does, they're even more true today. They're becoming more true, not less true. And so let's look at some examples now about how that exactly is. I pulled out three really core pillars. I went through a transparent marketing, pulled out three core pillars that were true then or true now. And let's see what they mean in the AI era. And this is the first one, it's clarity. And there's a direct quote from Flint, I'm sure you've heard it many times. Clarity trumps persuasion, don't sell, say. And anyone who started working with AI, that's the same thing that we're seeing going on with AI. So first I'm gonna ask Andy too, but first I wanna start Flint. We've got an example here of how have you been using AI to help with clarity? [00:15:09:06 - 00:15:44:07] Flint McGlaughlin So, you know, we were excited about what I meant for our research program in 2017 we announced our AI intentions in 210 days before chat GPT we launched our research cohorts testing AI. What you're looking at is the product of what was developed in those research cohorts. And in this case, here is an in market example this is the Hilton hotel. They have an event business. They have an agent delivery system like you see for every event. It looks like a chat bot, and you certainly can have an interactive conversation but it goes far beyond that. [00:15:45:08 - 00:15:59:01] Flint McGlaughlin It is trained so clearly on all of the detail of what it takes to produce a wedding at their venue, but you can ask detailed questions, you can get pricing estimates it will run calculations for [00:16:00:07 - 00:16:06:11] Flint McGlaughlin It will honestly give you a sense of what you need to pay for for you, even speak to sales. [00:16:07:12 - 00:16:10:09] Flint McGlaughlin And I can tell you that it is in market producing. [00:16:11:10 - 00:16:15:08] Flint McGlaughlin For those of you that know investment alpha results stellar results. In fact, [00:16:15:08 - 00:16:15:16] Flint McGlaughlin the [00:16:17:18 - 00:16:27:03] Flint McGlaughlin last conversation I had they asked me if we could create an AI system to automatically contact the people because there's so many leads coming in. And I'm having a hard time keeping up with them. That's [00:16:27:03 - 00:16:29:13] Flint McGlaughlin That was a real conversation [00:16:29:13 - 00:16:54:10] Flint McGlaughlin call. And I wanted you to be very clear, I'm not selling an AI platform when I say this, I'm selling an approach where we bring transparent dialogue into the conversation. And one more thing before Andy comments on this. The real value is not the leads. It's that every conversation is being analyzed by another set of conversion agents who are discovering, oh, this is what really matters to my prospect. [00:16:55:13 - 00:16:56:06] Flint McGlaughlin Years ago [00:16:56:06 - 00:17:01:07] Flint McGlaughlin I said the goal of a testing program is not a lift, it's a learning. I said the [00:17:02:07 - 00:17:13:12] Flint McGlaughlin climactic moment in a testing and marketing program is when you have successfully modeled the mind of the decision maker of your prospect as they view your offering. [00:17:14:16 - 00:17:26:00] Flint McGlaughlin Now we can do that from analysis of robust dialogue and at a speed we could never do before it's transformative. I'll step down and get out of the way, Dan or Andy, fire away. [00:17:27:08 - 00:17:28:03] Daniel Burstein Yeah, I'll mention [00:17:28:03 - 00:17:37:18] Daniel Burstein they're just great writers, Maxim, that I use as a writer. It's called Show Don't Tell, I'm sure you've heard of it before. And when I see this example when I hear Flint talking about it, and we talk about clarity, that's what I see too. [00:17:38:21 - 00:17:58:16] Daniel Burstein Before, just with a landing page, we'd be like, let's try to prove it. Let's tell them, tell them, tell them that's that sell part. And clarity trumps persuasion is, with something like this with an AI tool, you can just kind of show it and walk them through it. But this is when they're on the page, right? And then we also have to do some of those things. We have to do two things on the page as we talked about, Andy, right? We need [00:17:58:16 - 00:18:17:10] Daniel Burstein get the AI or Google or whatever it is to get them to the page. And then we need to also convert them on the page, right? We're doing those two things. So when you audit content for humans, I know you're looking for things like scannability and relevance, of course. But when you audit for machines, what are some of the top clarity failures you're seeing, and how can we overcome those? [00:18:18:15 - 00:18:47:10] Andy Crestodina Well, the example you have in front of you is classic. And it's a mock-up that I made for a piece once. And I teach this sometimes in conversations and in events. But the point is that specificity is critical to machines. And by the way, a subtle point. If you listen to Flint talk about this, he'd simplify it. He just reduces it to the word machines, which is very, very helpful because there's still too much of this search, traditional search, versus AI search. And which one? Oh, this is good for that. [00:18:48:17 - 00:19:08:10] Andy Crestodina Clear your head of that. And just think about machines, because there really isn't a difference between search engines and AI. They're all merged. There's no separate models. Chat GPT searches. Google summarizes. Just a quick side note. I have since been hanging out with Flint more. I just use the word machines. What you're seeing here, though, is an example of how [00:19:09:11 - 00:19:24:18] Andy Crestodina the lack of specificity and vague language-- Flint sometimes calls it ambiguity-- hurts you because it doesn't help the machine know specifically what you're doing. So specificity correlates with conversion. And this is an example of that. [00:19:25:18 - 00:19:28:24] Andy Crestodina And by the way, anything that I say that's good for machines, [00:19:30:06 - 00:19:44:11] Andy Crestodina I would never do something strictly and solely for machines. I'm doing it also for the human visitor. If you're doing anything that's just for a robot, you're probably not doing good marketing. You're not thinking about the market. I'm not sure you should be called a marketer if you're only thinking about a machine. [00:19:45:16 - 00:19:49:06] Andy Crestodina Having said that, I'll answer your other question there about what am I seeing. [00:19:50:11 - 00:20:05:09] Andy Crestodina If you don't mind, I'll share. This is an example. The term I use is data-driven empathy. And that, to me, is just a philosophy. It's like, how can I get better insights into my buyer? This is a vibe-coded [00:20:06:17 - 00:20:29:10] Andy Crestodina tool that looks at buyer signals. It's automatically pulling in sales call transcripts not available to everyone in every industry. I totally get that. If this isn't-- it's just an example of how you can get closer to your buyer. Even if you don't have access to sales calls, you can get other sources. But what it shows me is, literally, based on hundreds of calls and this number increments all the time, it's pulling in transcripts continuously. [00:20:30:16 - 00:20:33:09] Andy Crestodina What percentage of calls had which considerations? [00:20:34:17 - 00:21:01:15] Andy Crestodina And because our buyers are changing now, is that going up or down? So there are ways now. We can use AI and tools to better understand our audience. My more common methods are about understanding the buyer interesting inputs and insights that help me generate detailed personas which must be checked. I don't trust AI, per se. I believe in close human [00:21:02:16 - 00:21:08:02] Andy Crestodina observation of these things. But that little buyer signal is interesting because it'll show you, or I can see quickly, [00:21:09:04 - 00:21:21:20] Andy Crestodina where the clarity needs to come in. For us, pricing's at the top. Pricing transparency, 100%. What are the other things? Oh, here they are. What is the relative priority of those? Oh, here it is. How are those changing over time? Oh, now I know. So [00:21:23:04 - 00:21:45:06] Andy Crestodina this is an interesting moment that AI comes along with after COVID and a lot of us have sales call transcripts, fantastic insights. What do they need? It's not what I wanna say, it's what they need to hear. And I can know that I'm coming close to that. Not for all buyers. It's that these are super diverse populations. Often the visitor to the site is not the only person on the buying committee, but still, [00:21:47:04 - 00:22:35:19] Andy Crestodina if you run that analysis and then look at any of your pages, you'll find that probably some of these blocks should be reordered and the headers and body text should be more specific, adding more clarity, aligning better with the buyer. Don't try to guess all their prompts. That's impossible. And besides, it's not one prompt, it's a full conversation. But we need to better understand their moment of truth and to help them right then. And by the way, Flint's point's critical. Help them get out of your funnel and get off of your website if they're not qualified. Be explicit about what you don't do. It turns out, I don't know if you saw that, this is one of the most common questions. What don't you do? I don't want unqualified leads. Why wouldn't I try to? I don't want all the traffic in the world. No bottom line marketer. I'm gonna slow down before I... [00:22:35:19 - 00:22:51:22] Flint McGlaughlin with what you just said, I remember this conversation we had with the earlier example of the medicine for the Hilton and the head of sales, Vice President Sell said to me, our leads are far more qualified now and they're closing easier. [00:22:52:22 - 00:23:04:08] Flint McGlaughlin Which is critical because they couldn't handle the influx without, you know, that type of pre qualification. And that just affirms what you've just said, Andy, which is, you know, if we do this [00:23:04:08 - 00:23:11:07] Flint McGlaughlin right, the lead that comes in is worth the energy you must have asked in order to bring them to the next step in the journey. [00:23:13:04 - 00:23:21:05] Andy Crestodina Rankings is an input for traffic. Traffic is an input for conversions. Conversions is an input for qualified leads, which is what we're supposed to be doing here. [00:23:21:20 - 00:23:35:12] Andy Crestodina Play plan revenue, that's why marketers have jobs. That's why agencies have clients. That's why you have a webpage. That's why you're on the internet. That's why you have a business. If it's not a value offer, why wouldn't you say that you don't do that? [00:23:36:13 - 00:23:42:12] Andy Crestodina You're just making the internet noisier. You're making your marketing and AI systems less efficient. [00:23:43:07 - 00:23:50:03] Daniel Burstein Well, let's chat, you mentioned Andy, that's why we have a web page. So one of our attendees here is challenging that. And I wonder, I wanna [00:23:50:03 - 00:23:50:11] Andy Crestodina I'm good. [00:23:50:11 - 00:24:14:03] Daniel Burstein your perspective and Flint's perspective about that, should we even have web pages? So Brett asks, can you help with the context on what is the medium of conversion today? Where does conversion occur? Now the person no longer does the search, AI is doing the search, and the conversion seems to happen on AI without the brand's web page or website in view. Do we still need web pages? Do you have some thoughts [00:24:14:03 - 00:24:15:11] Daniel Burstein that, Andy, and I'll ask Flint next? [00:24:16:09 - 00:24:25:23] Andy Crestodina yeah. Your domain is your best hope of training AI to understand everything it needs to know about your brand. [00:24:27:01 - 00:24:32:04] Andy Crestodina I'm a Gen X self-taught marketer who started in 1999. [00:24:34:11 - 00:24:58:15] Andy Crestodina So I remember when apps came along and people said, "Those apps, why do I need a website?" Social media came along and people said, "No, either social, why don't you need a website?" AI comes along and says, "Really, what is your hope of training AI to know all the key points who align with all the decision criteria if you don't publish it?" It's the only place on the internet where you have unlimited time and space to go into all the details about your brand. Is it the mechanism for conversion? Maybe. [00:24:59:24 - 00:25:32:01] Andy Crestodina And weirdly, on so many sites I look at, I live in analytics, homepage contact thank you is the most common path on these websites. Like that's, they're just there to convert. And the conversion rates in visitors come from AI is extremely high. I'll share the data, we don't need to get into it, but why would AI conversion rates be higher? It's because AI believes you're a safe recommendation. Why would it believe that? It's because you published all the best answers. It's not the only source of information about your brand. But if you're a marketer, how do you, [00:25:33:05 - 00:25:44:01] Andy Crestodina where do you plan to teach AI? Where do you hope to train AI to be a sales rep? That's your new goal. Train AI to be a sales rep for your brand. If you don't have a site, [00:25:45:19 - 00:26:18:23] Andy Crestodina you're just like tying both hands behind your back. I don't know why you wouldn't do that. Even if you deprioritize the visual experience for visitors, but you deprioritize traffic, you prioritize the use of browsers, maybe the use of browsers will go down fine. We'll care. I just want qualified leads. And if AI recommends me, and if the person converts within the model, cool. But would it recommend me if I didn't share everything there is to know about what we do, for the value of it, we help, we don't help. I don't know how, how are you going to put all that on Reddit? Weird idea. [00:26:19:17 - 00:26:22:10] Daniel Burstein do you think, Flint? Do we still need web pages? [00:26:22:24 - 00:26:33:23] Flint McGlaughlin you. So, I'm going to start with the question of what is your website. Your [00:26:33:23 - 00:26:38:18] Flint McGlaughlin website is the brand truth that will become the canon by which you are evaluated. [00:26:39:19 - 00:26:45:14] Flint McGlaughlin But I would add to this, a new thing, a new way to think. I remember when [00:26:46:16 - 00:28:07:07] Flint McGlaughlin the iPhone came out it was a smartphone and it made us think of a new idea about flip phones they were dumb, and now there's a smartphone that actually, you know, can do so much more for you. The average website today is a flip phone where we're going, and I'll be teaching on this very soon, are to what I would call intelligent web pages. That is so important. I do not mean software driven web pages. Software is not intelligent. It's a deterministic tool. It is dumb. It takes forks in the road and makes decisions based on pre program paths. Intelligence is a power that is non deterministic. And what would happen if your web page was being re organized based on the nature behavior or source of traffic in order to put exactly the right message in front of that buyer, and what would happen if when an LLM comes to read this intelligent web page realize it's an LLM. And so it leads with a truth canon and gives the LLM what it needs. When you think about intelligent web pages remember something. The days of optimizing for a human. It was hard enough we broke them into segments that day has changed forever. Every website must be optimized for two evaluators, the human and machine. [00:28:08:10 - 00:28:19:21] Flint McGlaughlin And so, I would say Andy's 100% right. But for many of you on this call, I would start to wonder, how can I build an intelligent web page because you're going to see levels of performance, I'm already seeing [00:28:20:24 - 00:28:28:02] Flint McGlaughlin that you just could never get with a static message that you're trying to use to appeal to everyone. [00:28:29:23 - 00:28:30:07] Flint McGlaughlin Dan [00:28:30:07 - 00:28:30:24] Speaker 4 Yeah. [00:28:30:24 - 00:28:32:10] Flint McGlaughlin you. Yes, sir. [00:28:33:06 - 00:28:33:12] Daniel Burstein Yeah, [00:28:33:12 - 00:28:34:11] Andy Crestodina There's a... [00:28:37:04 - 00:28:45:21] Andy Crestodina Briefly. There's an AI audit tool now offered by Cloudflare with a step that says, have you introduced your agent to their agents? [00:28:46:24 - 00:29:06:12] Andy Crestodina A to A marketing. It's like a checkbox. It's like, oh no, you haven't done this yet. So what they're talking about is what Flint's talking about, which is not just building a site, but building a site that has its own agent that has ingested all of your best content. So that it has, so it's conversational for humans, but then also can, where these other agents can talk to your agent. [00:29:07:21 - 00:29:08:21] Andy Crestodina Easy for me to imagine. [00:29:09:13 - 00:29:14:05] Flint McGlaughlin it's, you know, it's possible now to start creating those experiences. [00:29:15:12 - 00:29:24:05] Flint McGlaughlin We're releasing them even now as I speak and we're building a toolset for creating intelligent web pages that will share at some point. But I think, [00:29:25:05 - 00:29:27:02] Flint McGlaughlin I think Andy's put his finger on the [00:29:28:03 - 00:29:35:14] Flint McGlaughlin on the most important piece here, the web page becomes more important in today's world. Back to you, Dan. [00:29:36:09 - 00:31:35:21] Daniel Burstein Yeah, so big, if you, one word to take away from all this, obviously we got into a lot of the details. You can go back to the recording, we'll be sharing the recording after this in the AI Guild Circle community. You can go back and see all the details. But the one big word to take away, to take away one thing, is that clarity element, right? So whether it's that intelligent web page, you're able to be clearer because you're more personalized, whether it's a human or an agent. Or on your static page, like Andy's got this great example on the screen. Rooted in research is a heck of a lot less clear what they even do than laboratory testing and safety done right. Now why, why does this come about? Because as marketers, we try to be creative and clever, and we do that before we're clear. So start with clarity. But clarity doesn't really matter a whole heck of a lot if nobody believes you. Right? So next up, let's talk about credibility. Right, and you can see there's a great quote there from Flint from the original Transparent Marketing. I'll just mention briefly one of my own experiences that anyone who's been in marketing and advertising might have had as well. My early days as a copywriter just starting out of college, I would sit in these meetings with these clients. And they would come up with all of these crazy claims they wanted to make about their products and their services and all that. And even as the incoporator, I would think, you could, back then it was newspaper ads and stuff, right? We have the budget, we can say whatever we want in the newspaper ad. Why would anyone believe this? Now fast forward to today, that's true still for humans and especially for AI. AI probably has raised the bar even more so because it's more skeptical than just the average consumer out there. So let's start talking about credibility and what that means today. So Flint, I wanna start with you. When we were first talking about this, I was asking you, what makes a website more credible to an LLM? How can a company structure its messaging and proof and content so the model is more likely to understand it and trust it and recommend it accurately? You had brought up a specific company, Precision Tax. And I wonder if you could talk through, here what we're seeing on the screen is some of the scores that Mecklabs AI has given Precision Tax or some of these things. I wonder if you could talk through some of the things they were doing on the website we can learn from that so we can have more credibility for the LLMs and the humans. [00:31:36:06 - 00:31:43:10] Flint McGlaughlin To do that I'm going to approach the issue from an inverted perspective so bear with me this will suddenly make sense. [00:31:44:10 - 00:32:09:24] Flint McGlaughlin When I asked people about the value proposition and they apply our methodology and they look for a fact that will allow someone to conclude, I can't get this anywhere else, just like I can get it from you in parentheses. Very important because until you have that answer. You've sold the category but not your business. When I asked that question the most common answer I hear is because we've got the best customer service. [00:32:11:05 - 00:32:17:11] Flint McGlaughlin And I immediately discount the, the answer heavily. First of all, I can't experience that till I have your product. So how [00:32:17:11 - 00:32:21:11] Flint McGlaughlin really going to help. Secondly, everybody says it so how do you quantify. [00:32:22:12 - 00:32:46:13] Flint McGlaughlin But I'm going to share with you the time when that answer actually produces value proposition force, and when it produces force which is something we can measure with our science, that force will influence not just the human but also the machine. So, I haven't been here for a while. But Dan if you're screen sharing. Why don't you go to precision tax calm right now. [00:32:47:14 - 00:32:54:02] Daniel Burstein All right, I'll pull that up here. While I'm doing that, I'll just pull up. We got a little overview of some of the things they were doing. Okay. [00:32:54:02 - 00:33:13:18] Flint McGlaughlin fine. Okay, so looking at the slide. I want you to notice and by the way they're, you know, they're, they're a category leader rapid growth etc etc. Those, when I first met with him I personally research the trust pilot and the Google rankings Dan can you make that bigger tell us how many Google rankings they have that are five star. [00:33:14:16 - 00:33:24:19] Daniel Burstein There is a lot. Let me see if I can try to zoom in here. I'll just pull up. I'll pull it up in the separate browser. Can you see my browser where I'm pulling up Precision Tax really? [00:33:24:19 - 00:33:26:11] Flint McGlaughlin could see it. Yes. There's [00:33:26:11 - 00:33:26:14] (Blank Audio) [00:33:26:14 - 00:33:26:18] Daniel Burstein All right. [00:33:26:18 - 00:33:33:02] Flint McGlaughlin versions of their homepage so you might see a different version wants to see what this one is but the point I want to make for you. [00:33:36:14 - 00:33:37:07] Daniel Burstein There we go. [00:33:37:07 - 00:33:39:05] Flint McGlaughlin right here. Click [00:33:39:05 - 00:33:39:06] Daniel Burstein We [00:33:39:06 - 00:33:40:13] Andy Crestodina Hundreds of reviews. [00:33:40:13 - 00:33:41:18] Daniel Burstein on Trustpilot. [00:33:41:23 - 00:33:42:16] Flint McGlaughlin and the [00:33:42:16 - 00:33:43:01] Speaker 4 Yeah. [00:33:43:01 - 00:33:55:22] Flint McGlaughlin numbers to and I honestly at the time I quickly determined there was no other tax service in the country with as many positive reviews. I also determined Andy that they didn't gain the system, they earn them all. [00:33:55:22 - 00:33:55:23] Speaker 4 Hmm. [00:33:55:23 - 00:34:12:06] Flint McGlaughlin And then I didn't know that I could actually believe it. Look, that's not a diverse audience, you could argue about the top of the amount of seriously like let's be realistic, you know that's not a diverse audience maybe that picture is in the right picture and yet this page killed me. It's hard to improve the conversion rate. [00:34:13:06 - 00:35:33:13] Flint McGlaughlin So I got deeper and I wanted to know, why is it that you're getting so many of these five star reviews because if I can't find a psychological reason I don't trust the mathematical reason I needed to find a cause and effect relationship I think this is just a hypothesis I discovered what it is. When you talk to them, even if you can't afford their services, they will try to help you, one way or the other, find a way out of your situation. They'll take extra time with you. They'll show you what you need to do next. And they've done it in a way that's so constructive that person after person is logged on there and spoken highly of them. Now, a marketer could have said that extra time is not worth it. This is an unqualified lead get off the phone and that's what your sales center will say that's what your typical you know manager of a cell of a call center is going to say to you. But in this case, the customer service and the customer first mindset produced a signal set that is the trust pilot review numerical that is quantifiable we quantify their customer service and we brought the metric into the front part of the conversation. Now a lot of folks put this stuff on your site, but very few can put this on their site and it'd be the highest possible number in the category. Can you see the difference. [00:35:33:13 - 00:35:34:12] Speaker 4 Mm-hmm. [00:35:34:12 - 00:35:43:01] Flint McGlaughlin this will both impact the machine evaluator and it will impact the human evaluator. And that's the point. Back to you, Dan and [00:35:44:01 - 00:36:11:02] Daniel Burstein questions we're getting in the chat, we're gonna share along with the recording, we're gonna share the slides in AI Guild after this event. And if you look at the bottom of the slides, there's a source that we're gonna add, there's a source that links, you can look at the page yourself, you can look at the article, whatever you wanna look at. All right, but Andy, you had given this beautiful defense of the website, the webpage just a few moments ago. I love it. It kinda reminded me of Patton standing in front of the American flag. Who are we? I love [00:36:11:02 - 00:36:26:03] Daniel Burstein should proof live so it's actually retrievable? We say the website, but is it on one page? Is it many pages, FAQs, schemas, documentation hubs? What are your thoughts there? I hope that you shared here what you're doing. [00:36:26:15 - 00:36:34:10] Andy Crestodina Yeah. Well, the previous example is just as good. So that said, America's most trusted ally, and right below that, was Trustpilot. [00:36:37:08 - 00:36:48:07] Andy Crestodina claim was supported with evidence nearby. So that idea is that if you say something, and that page you looked at, I think it was a good example, because often, [00:36:49:08 - 00:36:54:07] Andy Crestodina trust in evidence and proof points and testimonials and outcome data and [00:36:55:18 - 00:37:00:02] Andy Crestodina case studies and success stories are the only differentiated thing on the page. [00:37:01:02 - 00:37:04:20] Andy Crestodina Go look at your page and just say, could a startup born yesterday write this same page? [00:37:06:06 - 00:37:40:08] Andy Crestodina If so, you didn't, probably you haven't supported your claims, because when you add all those supportive claims, then that often is the thing that both AI needs, that visitors are looking for, that makes the story believable, and is the only thing different about what you're saying versus everyone else. Literally, on that precision tax page, the majority of the pixels were, the people, diversity or not, that's them, they're the only company with those people. They're real people, that's who they are, it's different. Not everybody has that picture. So, [00:37:40:08 - 00:37:42:02] Flint McGlaughlin they're in Idaho, right there. [00:37:42:02 - 00:38:06:04] Andy Crestodina great, yeah, whatever, yeah, that's who you talk to. That's who you, and I think there's some dots that could have been a little bit better connected than that, but the point is, the strongest pages, this is a big statement, and I shouldn't say things this strong when I'm on a call with Flint, but answer evidence, answer evidence, answer evidence call to action. That's sort of how it works. So, a common question in my world, who will I work with on this project? [00:38:07:07 - 00:38:24:24] Andy Crestodina That's one of the top questions, I have my buyer signals lined up. I wanna back that up, right? And so, how do I do that? I show us transparency, these are the people you work with, and not just are you gonna work with them, but are they good? Yeah, and by the way, I need to update that number, it's more, it's like 8.5 years, average 10 year at my company now. [00:38:26:16 - 00:39:05:13] Andy Crestodina 58% of my team has been here for five years or longer. So, I'm not just gonna slap that somewhere else, if you click to learn about our team, that's where I put that message. So, whatever claim you just made, I recommend putting the support for that claim nearby to make it easier for AI. There's a very prescriptive tip for this in the use of case studies today. If you have a case study on, a lot of people do this, homepage, success stories, case studies, you got a page block, it's got three case studies. Does that little blurb on the homepage include the juiciest nugget or the best quote or the strongest statistic from the case study? If not, you're making the machine go dig around. [00:39:07:00 - 00:39:30:16] Andy Crestodina So, the case studies, when they appear as snippets or blurbs or as part of a page block, make sure that that is front loaded with the reasons to believe right in there instead of just being generic. This goes far beyond what SEOs think about, which is EEAT, totally insufficient now. Oh, I added the, I added EEAT. That's a checkbox, you're a checkbox marketer. You're gonna get replaced by AI. You didn't think deeply about it. [00:39:31:20 - 00:39:47:01] Andy Crestodina The evidence should support the claim that it's next to and claims should be supported with evidence nearby. So, that is not common advice, but I think that is how sales works. That's how biopsychology works. That's how convergent optimization works. [00:39:48:06 - 00:40:20:03] Andy Crestodina Important point in Flint's example right there is that those were third party signals. What did they do? They pulled the things from the rest of the internet onto their page. But AI is reading those anyway. So, I didn't mean to overstate the importance of websites because AI learns about our brands from many sources. But whenever there's sparkling evidence somewhere else, you wanna put that on your site as well so that you're training the system to believe that you're a trustworthy recommendation to make. It's almost like thinking about the psychology of AI itself. [00:40:21:09 - 00:40:25:16] Daniel Burstein Yeah, and I do like your defensive web, please, because there's always been the on page and the off page that we can do. [00:40:25:16 - 00:40:31:01] Daniel Burstein this idea of like, okay, the off page, let's do that. That's gonna be hard. Let's just get rid of the on page, right? Like [00:40:31:01 - 00:40:31:13] Andy Crestodina Yeah. [00:40:32:16 - 00:40:33:04] Andy Crestodina There was [00:40:34:08 - 00:40:43:11] Andy Crestodina a comment just now that said they talk about EEAT a lot in the SEO firm. I think that's what SEOs should be talking about is, hey, did we make any unsupported claims? [00:40:44:09 - 00:40:47:12] Andy Crestodina We're gonna EEAT, think about, did we make any unsupported claims on this page? [00:40:48:11 - 00:41:04:14] Daniel Burstein Well, I wanna ask you to direct me, and I'll ask you this next one. Because you did talk about, you got some numbers here as an example. You talked about case days and all these things. So let me ask you directly, what forms of proof travel best through an LLM response? Right, there's numbers, there's policies, third party citations, customer constraints, case studies. What do you think, Andy? [00:41:07:01 - 00:41:07:10] Andy Crestodina Well, [00:41:09:14 - 00:41:45:13] Andy Crestodina briefly, I have a friend who has a company called Case Study Buddy. And when people ask him, can we use anonymous testimonials, he says, actually, that can build credibility because it shows that you are protecting the confidentiality of your clients. I think that stories are generally more compelling than data in human psychology. And that case studies, especially like a video case study, or video testimonial can be extremely powerful. But it just depends on the claim. If you say, you know, fast delivery, and you show the percentage of your deliveries that are on time, that supports that claim. So, different types of claims probably need different types of evidence. The goal is not to use testimonials. The goal is to support every claim. [00:41:46:21 - 00:41:50:15] Daniel Burstein What are your thoughts? What forms of proof travel best through an LLM response? [00:41:50:15 - 00:41:53:16] Flint McGlaughlin all, we control the chronology in the mind of the geography of the page. [00:41:54:17 - 00:41:54:23] Flint McGlaughlin This [00:41:54:23 - 00:41:55:03] Speaker 4 Hmm. [00:41:55:03 - 00:42:00:18] Flint McGlaughlin the proximity of your evidence in proportion to your claim is critical. [00:42:01:24 - 00:42:17:18] Flint McGlaughlin Secondly, when you make quantitative statements like we're fast, customer service is good, and you can't back them up with qualitative evidence. You let a third person back to Emily Dickinson make the claim about you, quote someone else, [00:42:18:16 - 00:42:22:02] Flint McGlaughlin because we don't trust you when you brag about yourself. [00:42:23:12 - 00:43:07:01] Flint McGlaughlin If indeed your marketing doesn't sound like a conversation you could have with someone you just met after work, it's not right yet because you're using rarefied language and probably language patterns that don't produce trust, and you say things. You wouldn't walk up to someone, say at a bar after work, and say, "Hi, my name is Andy. I'm the leading expert in AEO and conversion optimization." Never in the world would you say that. Now, Andy, you wouldn't say it on your web page, but marketers all day long say that, I'm going to give you a very technical scientific term, use that crap, and then go out to interact people and use a whole different set of language patterns. [00:43:08:04 - 00:43:32:03] Flint McGlaughlin Be careful, guys. I would also say to you this. I've noticed now that these landing page builders, such as Lovable or Replit, or just fill in the blank, will tend to produce a page and even put fake testimonials on the page for you. It's very easy to run with what they've already given you. Marketers do not cross that line. [00:43:33:02 - 00:44:00:17] Flint McGlaughlin The minute you start using testimonials that you didn't earn, the minute that you start using case studies that aren't real, you have violated trust in a way that is so profoundly disturbing, you can't yet recognize the downstream consequences, both to your brand and to the rest of us out there who are struggling to get hurt over the white noise. It's very easy now to launch your page with testimonials that aren't real. Beware of the trap. [00:44:01:24 - 00:44:10:00] Daniel Burstein That's a great point, Flint. And I wanna ask Andy, if you've come across any good way to find these reasons to believe, right? Cuz here [00:44:10:00 - 00:44:10:07] Speaker 4 Hmm. [00:44:10:07 - 00:44:42:21] Daniel Burstein an example of what you're talking about, how the pages should look. Because as Flint talked about, right, we get this AI output, we throw it up there. And then sometimes I have pushed marketers in that before. I'll get pitches for a marketing Sherpa case study. I won't choose to pitch even if it's an okay pitch, cuz they have an obviously fake testimony on there. And ask them why they do that? And they're like, well, it's hard to get testimonials, or it's hard to find these reasons. And so I know we've mentioned you've worked with many, many companies across many different industries. Is there any advice or something that's worked well on how to find these reasons to believe? Any version of that looking at the sales logs that you mentioned or the sales conversations in [00:44:42:21 - 00:44:43:01] Speaker 4 Hmm hmm. [00:44:43:01 - 00:44:43:10] Daniel Burstein company? [00:44:44:14 - 00:44:50:24] Andy Crestodina Well, humans are generally bad at gap analysis. AI is very good at gap analysis. I recommend using a prompt I'll share. There's [00:44:52:07 - 00:45:00:01] Andy Crestodina an article with a link that has that prompt in it. This prompt will look at your pages and give you a list of every unsupported claim you made. [00:45:01:09 - 00:45:10:05] Andy Crestodina And then your job is to support those claims, the ones that add proof to the most important claims to your visitor, to your audience, to your buyer, your prospect. [00:45:11:06 - 00:45:11:20] Andy Crestodina The [00:45:12:21 - 00:45:53:21] Andy Crestodina power of testimonials, and this is where marketers are completely safe in the AI era, is that you have to go offline to get them. If it's the next job of marketers is sort of an offline investigation. Go find those things and put them on the internet because that happy news is in the mind of your super fans. You have to elicit that and pull that in and put that on the web. Offline investigation is one of the things that marketers must do now in this era because it's not on the, you know, the internet's not finished yet. You have to go find those things and put them out there. How to get them, I think there's lots of tips for this. They definitely predate the AI era. Social listening, events, customer events often do this. Follow-up calls. [00:45:55:00 - 00:45:58:24] Andy Crestodina I'm friendly with John Jans, who's a Flint fan. [00:46:00:07 - 00:46:27:09] Andy Crestodina Yeah, John. And John talks a lot about what Flint talked about a minute ago when it comes to eliciting reviews, testimonials, very similar, social proof, which is just providing great customer service. And that follow-up, that bit of extra service, this is not normal marketing advice, but in what ways did your service surprise and delight? What was the feedback of that? What did you do at the end to follow up and learn and figure out how you could do better? [00:46:28:12 - 00:46:48:12] Andy Crestodina Every project we do, we do a post-follow-up call to confirm that we met expectations and they're happy. That's when you get your testimonials and reviews is when you follow up on your service to discover how you could do better. You didn't do the call to get a testimonial. You did the call to get feedback on how you do your work. And testimonials just pop right out of those. [00:46:49:12 - 00:46:52:14] Daniel Burstein I like it, it's also doing the right thing and doing the right thing for the customer. [00:46:52:14 - 00:46:53:13] Andy Crestodina Yeah, service. [00:46:53:13 - 00:47:31:19] Flint McGlaughlin do that, Dan, I recommend that you take those testimonials and you categorize them in a spreadsheet based on the anxiety or friction someone would anticipate with your product. Did I pay too much? Is it going to be okay? Can I uninstall it without a problem later? Or can I cancel without it being? Once you have those broken into categories, instead of spraying them on the page and hoping, you use them quite strategically to produce the conclusion back to what we said at the beginning, you want them to arrive at. So, engineer the appearance of the testimonial along with its theme to support your conclusion set. [00:47:32:23 - 00:47:41:02] Daniel Burstein And one reason we're using testimonials was for credibility, but it's also the third movement we really wanna address here is anxiety, right? So [00:47:41:02 - 00:48:12:01] Daniel Burstein reason we also use it is to overcome that anxiety. A user might have, a visitor might have, should I hire them, what will happen, right? So Flynn, I wanna talk to you about anxiety. How is it, do you know, anxiety is many people watching know, some people might not, as part of the conversion heuristic we talked about, Flynn's got a patent for ten different heuristics. One of the most famous is a conversion heuristic, which talks about the different elements that affect conversion. Anxiety is one of key those. I wanna ask, how's anxiety the same and different for humans and machines? And what should we have on our websites to address it? [00:48:12:11 - 00:48:30:09] Flint McGlaughlin At some point, we should probably drop in a link to the science article on LLM displacement theory. Dan, it's on the megaloves.com research site and current research, because it will get into great on this and tell you even how to score your site so that it gets recommended. But I will borrow from that to tell you this much right now. [00:48:31:13 - 00:49:13:08] Flint McGlaughlin Anxiety in the human is typically very different than friction. When we first did the research in the lab 20 years ago, I had them clustered together, Andy, as to one piece, resistance, psychological resistance. But I realized that I could fix friction with a one-to-one correction, but anxiety had to be overcorrected because often it's emotional, it's irrational. In a human being, you must overcorrect friction. And yet you would wonder, well, a machine is not emotional, so how would the conversion sequence and its attention to anxiety produce any relevant help for me when I'm trying to get recommended by a machine? [00:49:14:15 - 00:49:21:11] Flint McGlaughlin And the answer is this. Machines don't experience emotions, but they experience conditions. I want to say that again. [00:49:23:09 - 00:49:36:24] Flint McGlaughlin You are thinking for the human evaluator about emotions. If you're thinking about the machine, you're thinking about conditions. Conditions and emotions work in similar ways in this particular context, and I'll give you an example. [00:49:38:04 - 00:49:44:21] Flint McGlaughlin A while ago, we talked about those testimonials not being fake, Andy, just a few minutes ago. [00:49:44:21 - 00:49:57:00] Flint McGlaughlin There's deeper truth than that than, say, not disappointing a human evaluator. The machine is looking for an agreement between the public truth and the brand truth. Did you hear me, [00:49:57:00 - 00:50:29:02] Flint McGlaughlin truth disagrees from the brand truth, the machine is going to determine that you may not be something they can trust, and it will experience the condition I'm going to call anxiety. I'll tell you why in a moment. Let's get to another example. If the machine comes to your website, it can't figure out, for instance, let's take a hospitality, because I'm talking to the leaders in the hospitality industry about this. I've been flying to the ranch from Washington in two weeks to discuss the impact on the national chains, and I can tell you what we've discovered. [00:50:30:09 - 00:50:59:06] Flint McGlaughlin Right now, you can go to a major hotel chains website, fill in the blank, I won't name them, and you won't be able to figure out what your room's going to cost. How many already know that? Because there's a resort fee they tell you about when you get there, and, oh, what about the parking that was an extra $35 that they didn't tell you? And because the public truth is as confused as the domain's brand truth on the website, so the machine says, I don't know what this costs. And the machine may make other conclusions based on inconsistencies. [00:51:00:08 - 00:51:16:04] Flint McGlaughlin Now, I mentioned, and I say to you a relational example, if my daughter introduces me to a young man that she's interested in, and I start to speak with him, and I start noting inconsistencies, there's my first clue that I need to be careful about trust. [00:51:16:04 - 00:51:53:22] Flint McGlaughlin maybe her emotions are very wrapped up in that new relationship, but I'm a father looking for those inconsistencies right now, because I've lived long enough to know how important they are. Now, let's go back for a moment. What is the machine's condition of anxiety? The machine must make a recommendation that is defensible. That's a key word. It's programmed in its algorithmic foundation to be able to defend the recommendation set. If it can't defend the recommendation set, then it doesn't want to make that recommendation, because it's in a condition of uncertainty. Can you see the parallel now? [00:51:55:01 - 00:52:53:08] Flint McGlaughlin As we bring consistency to the conversation with a human evaluator, as we bring it to the conversation with the machine evaluator, as we bring consistency to our internal truth as published in our own brand, and as we bring consistency to our external truth as published in the public domain body of content, the machine is able to reduce anxiety, and our recommendation becomes more defensible. And by the way, if you're on this call and you're a small business, you're in luck. Because I'm talking to gigantic brands that are terrified because they don't know how to get the public domain right. They're going to try to flood it with agentic content to drown out the inconsistencies because they can't take them down. Many of you don't have that problem. You can get a good, clean slate right now. And this is the thing that might surprise you. You might be an independent end in the same town as the Marriott, trying to serve the same customer and find that you get recommended and the Marriott does it [00:52:54:10 - 00:52:59:16] Flint McGlaughlin because you've got these pieces properly and they do not. Back to you guys. [00:53:00:16 - 00:53:17:13] Daniel Burstein talked about one specific website, so I'm gonna pull that website up again, the strategic business valuations and what you thought, or how you work with them to help improve anxiety. So let me pull that up here, so everyone can see it. And I'll just kinda, is there anything specific you wanna call out here about what they were doing for anxiety? [00:53:17:24 - 00:53:20:00] Flint McGlaughlin Yeah, so I do. [00:53:21:12 - 00:53:26:11] Flint McGlaughlin And I want to do it in a certain way and then get Andy's input because he has so many valuable insights. [00:53:27:23 - 00:53:38:18] Flint McGlaughlin But let me just say this to you. To really appreciate it, it's a small business. They do M&A work and they do business valuations. And their original website looked like a small business [00:53:40:01 - 00:53:40:20] Flint McGlaughlin and it, [00:53:41:24 - 00:54:02:23] Flint McGlaughlin in its own basic design, conflicted with the conclusion set. Are you hearing marketer? The conclusion set that you needed to draw about them, not that they were huge, but they were highly professional, that the service would be personal, that they had the depth of expertise. I could go on the list of things that needed to be concluded. [00:54:04:00 - 00:54:11:13] Flint McGlaughlin So in order to draw that conclusion, look at the website on the right. That was what they started with. Know [00:54:11:13 - 00:54:27:01] Flint McGlaughlin value, create your legacy. You could put that on a hundred different types of websites. They don't even belong in the same business category they're in. That's your first clue. It's wrong. Everything they're after looks like one of those templated WordPress sites in the old days, right? You know what I'm saying? They [00:54:27:01 - 00:54:31:14] Flint McGlaughlin just have the pieces. They're barely covered, but there they are. Hey, we're in business. We have a website. [00:54:32:18 - 00:54:52:08] Flint McGlaughlin When you think conversion optimization at the level they're trying to do, because of the nature of their ICP, they had it. And then let's go back to the original. They had to have a website that you'll notice, and notice this carefully. It still has, like the original scroll down, if you would, Dan, pictures of the principles. [00:54:53:10 - 00:54:53:16] Speaker 4 You [00:54:53:16 - 00:54:56:11] Flint McGlaughlin And so we're not pretending to be a hundred times bigger, [00:54:57:15 - 00:55:02:12] Flint McGlaughlin but these pictures themselves were optimized. Can you feel the difference? They're [00:55:02:12 - 00:55:02:22] Speaker 4 Yeah [00:55:02:22 - 00:56:04:05] Flint McGlaughlin They're genuine. I'm talking to these people. These are the real people. And yet it's ensconced within a context that looks highly professional and like there's depth. And if you click on the on the agent there who's working, not like a chatbot, it's depth of experience, understanding the industry, answering critical questions at a level you'd never get, and then quietly offering to help you even deeper with that 15 minute call is all locked together to produce a consolidated sort of impact on your thinking, because optimization doesn't occur on the web page. It occurs in the mind, and that's stuff on the web page. That's just zeros and ones. That's just signals. If we put the right signals in front of somebody and they observe them properly, they will draw certain conclusions. The conclusions will then power a decision or decisions and the decisions power the behavior. If the behavior doesn't change, it wasn't a real decision. [00:56:05:11 - 00:56:28:08] Flint McGlaughlin If you walk through that again, observations, signals, conclusions, decisions, behavior. The business we're in as marketers when we designed a web page is arranging our signal set to produce the right conclusions that will power the right decision that'll get the right behavior, which to speak crassly, Andy might be a deposit in your bank account, [00:56:29:10 - 00:56:34:18] Flint McGlaughlin but it starts way back there at the zeros and the ones. I'll leave it back [00:56:34:18 - 00:56:59:15] Daniel Burstein Yeah, I'll put on one thing, we'll get this correct, it slides out to you. The, what's labeled the after, this is actually the before on the right and on the left is actually the before. So we'll get that corrected out to you, but that simpler page was actually the before. Andy, let me ask you one last thing. You were kinda hinting at this before how you analyze all these sales calls. And what are some examples of helpful limitations that increase trust instead of lowering conversion, and how do you find these? [00:57:01:04 - 00:57:08:10] Andy Crestodina And Flint, I mean, that was both deeper and broader than the thinking of most marketers. [00:57:09:09 - 00:57:13:03] Andy Crestodina That is really the biggest picture perspective on it. [00:57:14:09 - 00:57:17:02] Andy Crestodina The method of using sales calls is just to try to [00:57:18:13 - 00:57:40:11] Andy Crestodina understand the buyer and be able to construct a page that is a visual hierarchy where their prioritized information needs to get met as their eyes move through these pixels. But that, I saw an example just the other day of a company in a similar category, not just valuations, but they do forensics, like accounting forensics. The visitor to that page thinks that someone in [00:57:41:17 - 00:57:43:13] Andy Crestodina their company is stealing. [00:57:44:21 - 00:58:09:15] Andy Crestodina And for that visitor, confidentiality is critical, but the page didn't mention that. So just imagine that person's prompt. Imagine that person's buyer journey. Imagine their top considerations and concerns and hopes and dreams and fears. If that word is in there and there's nothing in the marketing that aligns, why would, what did we say? Observations, signals, decision, behavior, [00:58:09:15 - 00:58:10:21] Flint McGlaughlin observations, signals, [00:58:10:21 - 00:58:15:10] Andy Crestodina signals, conclusions, decisions, behaviors. Thank you. [00:58:17:01 - 00:58:20:21] Andy Crestodina I learned so much from this. I hope that others have gotten as much from this as I have. [00:58:22:04 - 00:58:41:01] Andy Crestodina It's true in every case. We'll pick any category, any vertical, any niche, any service, any buyer's decision criteria. Watch your own, I don't like using myself as an example and I tried to never do that, and we're all datasets of one. But observing your own behavior, you'll see this. Like, [00:58:42:00 - 00:59:01:21] Andy Crestodina does it work? How soon will I get it? What is this thing? Is it gonna fit? Does it work with MySync? Whatever those things are. Any time that you didn't get matched with a brand, it's often simply because that brand didn't add I'll throw Flint's quotes back at him once again. High clarity, [00:59:02:24 - 00:59:11:07] Andy Crestodina high corroboration, low ambiguity. If that's all you do next is, and that touches across, I think a lot of what we just discussed, [00:59:13:11 - 00:59:14:10] Andy Crestodina you're on the right track. [00:59:15:04 - 00:59:46:12] Daniel Burstein We have a lot of advice, we got one minute left. I'm gonna ask both of you, but I'm gonna answer first. I'm gonna ask both of you, if you could tell everyone on the call, we talked about a lot of stuff today. What one piece, what one thing would you say them to do next? And I'll, let me say mine first, go to meclabs.com slash resource kit. We got some resources for you here, meclabs.com slash resource kit to help you put some of the stuff in action. It's gonna take you to an AI guild chat. We've got some links to additional resources from us, from Andy. With that, we have a minute left split. If you could give them one piece of advice, what should they do next? And I'll ask you next, Andy. [00:59:47:13 - 01:00:01:16] Flint McGlaughlin Marketers need to be profoundly focused on messages that will only be relevant a thousand years from now or would have been relevant a thousand years ago. To do that, every single conversion-focused message must produce [01:00:01:16 - 01:00:22:06] Flint McGlaughlin eight micro-yeses and they typically work in a linear fashion. Here they are. Number one, yes, I will pay attention. That's what your headline does. Number two, yes, I will engage. You got me for a split second, but can you get me now to engage with the content beneath or in the phone call, wherever that message unfolds? [01:00:22:06 - 01:00:36:06] Flint McGlaughlin Number three, yes, I understand. Yes, number four, yes, I believe. I can't understand, then I can't believe. If I understand but I don't believe, I still lose. Number five, [01:00:37:10 - 01:00:46:24] Flint McGlaughlin yes, I want this from you now. Number six, yes, I can't get this anywhere else, at least exactly like you deliver it, except from you. [01:00:48:08 - 01:01:09:01] Flint McGlaughlin Next, yes, I will trade. That's the filling out of that form. That's the exchange on the e-commerce side of their money. And finally, and you can't stop without number eight, yes, I will finish. You have to get all eight every time in your process to get a conversion. [01:01:10:07 - 01:01:19:10] Daniel Burstein Andy, you're our guest. Thank you so much for being our guest today. We'll give you the final word. We talked about a lot of stuff. One piece of advice, what marching orders would you give to all of us today? What should we do next? [01:01:20:00 - 01:01:26:04] Andy Crestodina Jeez, how do I follow that? Daniel, that's so unfair. Study value propositions. I think that's, [01:01:27:16 - 01:01:29:20] Andy Crestodina obsess over your buyer. Think [01:01:31:19 - 01:01:35:14] Andy Crestodina bigger, go back farther, and [01:01:36:23 - 01:01:45:10] Andy Crestodina let's not forget that even though it's digital, these are fundamental foundational marketing best practices that will always be true regardless of tech disruption. [01:01:46:18 - 01:02:04:02] Daniel Burstein Thank you, Andy. Thank you, Flint. Thanks to everyone for attending. Remember, you don't have to like now leave this and go it alone and do it all by yourself. Go into the AI guild circle community, collaborate, share your ideas, share your questions. We'll all help each other out. And then we'll get together again in November and learn some more stuff. So thanks everyone for joining. Have a good one. [01:02:04:02 - 01:02:07:03] Andy Crestodina Thank you, Flint. Thank you, Daniel. [01:02:07:08 - 01:02:08:09] Flint McGlaughlin All right, team. [01:02:09:14 - 01:02:10:15] Flint McGlaughlin Take care, everyone. [01:02:10:15 - 01:02:10:20]