← CMO Confidential
CMO Confidential
Chris Andrew | CEO Scrunch AI | Is Your New Best Customer an AI BOT - Welcome to the New Frontier
00:32:55
Loading summary
Transcript65 lines
- [00:00]
Mike Linton
The CMO Confidential Podcast is a proud member of the iHear Everything Podcast Network. Looking to launch or scale your podcast, iHear everything delivers podcast production, growth and monetization solutions that transform your words into profit. Ready to give your brand a voice then visit iheareverything.com welcome to CMO Confidential.
- [00:25]
Chris Andrew
The podcast that takes you inside the drama, decisions and choices that go with being the Head of marketing. Hosted by five time CMO Mike Linton.
- [00:36]
Mike Linton
Welcome marketers, advertisers and those who love them to Chief Marketing Officer Confidential. Chief Marketing Officer Confidential is a program that takes you inside the drama, the decisions and the politics that go with being the Head of Marketing at any company is in what is one of the most scrutinized jobs in the executive suite. I'm Mike Linton, the former CMO of Best Buy, ebay, farmers insurance and ancestry.com now this show is brought to you by Props, our title sponsor. Props is a storytelling as a service platform which creates hyper specific content for hard to reach customers. Using authentic stories and professional creators. These stories are designed to produce both awareness and leads. Check them out at Props Co. Now I am here today with my guest Chris Andrew and today's topic is your best customer is an AI bot. Welcome to the new frontier. Chris is the CEO and co founder of Scrunch AI, a company that optimizes your brand for AI search. Previously he was the Chief Product Officer at Hearsay and in his spare time he makes wine which I can vouch for as being pretty good. Full disclosure, I met Chris at the Marketing AI conference where his thinking on how AI might impact brands and marketers was an eye opener. Welcome Chris.
- [02:01]
Chris Andrew
Thank you Mike. Good to see you again. We spent five hours in our first chat. Hopefully very good.
- [02:07]
Mike Linton
It will be faster this time though. So let's go right to the main topic. Why will your new best customer become an AI bot? And this is like give us an overview of whole marketplace today and where it will be a few years down the road and write the AI bot into that story.
- [02:26]
Chris Andrew
Let's do it. Do you remember where when GPT launched? Let me ask you a question first.
- [02:30]
Mike Linton
Remember it was November of.
- [02:32]
Chris Andrew
Yeah, you're on it November 22nd. We're coming up on two years and I think what's crazy about the marketplace today is a lot of people don't know the behind the scenes story of GPT launching. The technology existed. They launched it really as a playful experiment to showcase the power of the model. It wasn't really meant to be a viral consumer tool. But GPT goes out and all of a sudden everyone's like, this thing is amazing. It gives me answers instead of a bunch of links. It thinks for me, which can be good and bad. What I think that did to the marketplace was it captured the attention of consumers and through the business world into a frenzy of AIs, changing everything. People start kicking off projects, initiatives, figuring out how to integrate AI into their technology stack. But I think the undercurrent behind the scenes was that it immediately changed consumer expectation. It went to a world where I expect to get an answer from this tool. It simplifies my experience in searching, in researching and investigating in the same way that Amazon kind of simplified the purchase experience. So I'm a big believer if you can simplify something for a consumer, you're onto something.
- [03:42]
Mike Linton
And I think this simplification is when I, when I used to get, or when I did do get search results, today, it's still homework. And I can click on any of the buttons and I can keep rolling all the way or any of the links. I can roll through multiple pages of links. But now I'm going to actually get answers like the five best restaurants in, you know, Buffalo, New York, or something. And those are going to be answers versus links. Is that right? And. And tell me how this evolves. Where now the customer is. The customer of the marketing is an AI bot.
- [04:16]
Chris Andrew
Yeah. If you think about your customer as being your customer. If your customer is using an AI bot to do that research, to ask the question of where to eat dinner, what are the best headphones to buy? What's the best minivan for my growing family? If I'm asking my assistant, and the assistant's going out and performing that search, I, as a marketer better be optimizing my brand to engage with the AI bot, which is a, which is a huge mindset shift to go from engaging with customers to engaging with this AI assistance that's doing the research for my customer. But I think the parallel, parallel is it's similar to what CMOs had to do for Google. I talked to a CMO and they're like, my entire website was built for the Google Crawler, not a human for the crawler. Now what? Now you need to build your website and your presence for an AI bot that's getting information on behalf of the consumer. So when we say your customer is an AI bot, it's because they're partnered with your customer.
- [05:10]
Mike Linton
But then also this takes the search tool out of your hands. Correct. If you're a marketer and I'm searching for those five restaurants. I'm not, I'm not going to the search links. I'm just getting an answer and then I'm going from there. So as a marketer, suddenly it's is this two cuts? One is I have to market to the bot, and two is my search power is decreased. Is that what's going on here? Really?
- [05:39]
Chris Andrew
Yeah, I mean, you know, Gartner came out with a report that they expect 50% of organic search volume to disappear in the next two years. That's pretty startling, right? You got, you got large brands that are partnering with Gartner for advice, and they're saying the front door to how you access customers is changing at a level of 50% of a shift. That's a huge shift. And I think it's happening because AI is infiltrating channels that consumers are already comfortable with. When I go to Google to do a search today, more often than not, I'm getting a Google AI summary.
- [06:10]
Mike Linton
Oh, yeah, yeah, yeah. Why?
- [06:12]
Chris Andrew
It's because Google's cannibalizing its own results because it sees consumers are preferring it. So you're absolutely right. It's changing your search strategy, it's changing who you have to market to. You have to be aware of how to connect with these bots, but it's not making search engine optimization irrelevant. There's still a search happening behind the scenes to choose what to present as an answer. So search is still a critical part of it, but it's the bot that's doing the search rather than the consumer.
- [06:37]
Mike Linton
Okay, well, we're going to circle back on this, but we also talked about the rise of generative AI and honestly the destruction it will cause. You've just mentioned search, but also you talked about the death of websites and the collapse of consideration. Yeah, I feel like we're talking about John Connor and the Terminator and I feel like I have to go back and watch the movie. Tell us what you mean by all about all of this destruction.
- [07:08]
Chris Andrew
Yeah, I think think about your own behavior. And I'm on frequently with CMOs and VP of Marketing who are using these technologies day to day in their own life, but haven't deeply thought through what it does to their brand, to their product and service. And so when we say the death or the destruction of the website, what I believe is that people are going to be a lot less likely to visit your website again. Their assistant is visiting your website. It's getting the information, it's bringing back a text summary. Maybe it's bringing back a link. But I'm so much farther down the purchase funnel by the time I'm ending up on an owned property of yours because my personal behavior is I'm just chatting with my assistant, I'm trusting the answers it's giving me. I'm a lot less likely to browse the web. So people still read newspapers. They kind of do it for fun. It feels good. It's kind of entertainment to flip through something. I think we'll see a shift in web browsing behavior towards I may want to browse for fun and kind of click around and look at things. When I'm doing research or making a decision or getting Advice, I'm asking ChatGPT, I'm asking Siri, I'm asking Google and it's giving me an answer. And I'm less likely to visit the website itself.
- [08:17]
Mike Linton
And this will be say, rather than going to a retailer and shopping a bunch of different winter jackets or something, I will say what are the five best winter jackets? And then it will be locked down into that jacket to go buy it. And I won't go to the like the Columbia website or anything. I will just go immediately to Amazon or wherever and buy this jacket. Is that what you think is going to happen?
- [08:41]
Chris Andrew
I think so. I mean also you think about that process of being on one of those retailer sites and looking at 50 different jackets and trying to filter by your gender, filter by size, filter by season. Let's start to think about all these things these models and assistants are going to know, right? So when Apple Intelligence releases on billions of Apple devices with GPT built in in a privacy secure way, it knows your texts, it knows your emails, it knows your calendar, it knows what time of year it is. So if I'm asking for a jacket and I'm sitting in New York in November, it's probably presenting a winter jacket for me, knowing my size, my tendencies, my past purchase behavior, that's a lot of information filling in quickly. And I might trust its recommendations in the same way I trust a friend who's a fashion aficionado saying, hey, this is the jacket for you. These are the five jackets for you. Less browsing the web to kind of zone in on that purchase.
- [09:32]
Mike Linton
So I think and eventually it will do in your mind if I carry this to the Terminator Extreme, it will do all the consideration for you and it might give you a couple of choices, but it will eliminate a huge part of the marketing consideration funnels. Am I hearing you say that right?
- [09:48]
Chris Andrew
Yeah, we've, I was in a meeting recently where someone said, this disrupts word of mouth for the first time. I was like, why? Oh, it disrupts word of mouth because, like, I know you ride bikes. I know you like theater. If I've got a question about something in those categories, I might come to you and say, hey, what are the best five productions you saw in New York last month? What bike do you recommend for gravel riding? I trust you because I know, you know, it's. If I'm able to know that these assistants and models know me and know my tendencies, the end of consideration is kind of a scary thing. And I actually think when people talk about the risk and the fear of AI, I do think there's an undercurrent of making sure we continue to think for ourselves. Right. And not completely outsourcing that thinking, but where simplicity can be introduced to complexity, consumers prefer it. So I do think we're going to see somewhat of an end of consideration because these assistants and models brings so much awareness to a purchase process.
- [10:44]
Mike Linton
That's. That's a great. That's a really interesting point. Thank you for that. Let's stay on the destruction theme, since it has so much to it.
- [10:52]
Chris Andrew
Yep.
- [10:53]
Mike Linton
You also stated when we were talking that chatbots can kill you. And what did you mean by that?
- [11:01]
Chris Andrew
Couple different things. I mean, you think about the existing behavior of me on Google, me on a Microsoft property, me on a meta property, asking a question. And behind the scenes, AI is doing the search. My AI assistant's going out that. That chatbot's going out there to get information for you. If the information isn't well presented, if it's inaccurate, if there are gaps in it, that AI assistant doesn't care. It doesn't go back to your website and say, hey, Mike, you're missing this. It just skips you. It just goes to the next website. Right. We're finding all these tendencies in terms of how do these AI models prioritize and value content? How do they want it structured? When there's a gap in your content, what is its behavior? Is it going out and finding the next thing, or is it going to the next level, to the next site on your. On your page to find the information? Speed is such a huge factor when you think about the dominance of Google search. Right. Google is so quick in giving us those links and that information. We're finding that if the chatbot can't get its information quickly, it bypasses you. It goes.
- [12:04]
Mike Linton
So you have to be structured, you know, like, you have to be structured for Google, you're going to have to be structured for all these bots and each of these bots is essentially different, or do they have similarities to them?
- [12:16]
Chris Andrew
There's some similarities, but they're different. I mean, I think for better or worse, Google has had such a stranglehold on search for decades. Right. Like that's the game. There's, there's some smaller players and people still invest there across Bing and Yahoo. But the difference here is we're talking about models from perplexity, from chat, from meta, from anthropic. Microsoft has its own models. I think we're going to see 5 to 10 models both for B2B buyers and B2C buyers that are relevant in the equation. And so you're going to need to be aware of the differences between them. I'll give you one example.
- [12:51]
Mike Linton
Do. That's great. Thank you.
- [12:52]
Chris Andrew
ChatGPT is 100% powered when it comes to these live web searches by Bing. Now, if you're, if you're a marketer, you probably haven't spent too much time thinking about optimizing search for Bing. Google's the game. I think this is kind of the dark horse in the game in terms of if Microsoft has been going after that search kind of title, that search leader position for decades. What if ChatGPT is the greatest volume of search in the future and all of a sudden Bing becomes the number one search engine because it's GPT that is powered by Bing behind the scenes to get the results. That's one example of something that I think a lot of marketers aren't thinking about right now.
- [13:31]
Mike Linton
So I have to deal with. I think you named at least five models and you know there's more coming. So let's say there's 10 models. We've already talked about the impact on search, which you're outsourcing, a lot of consideration. And you're also. It's not Link's homework anymore. It's now it's your choice. When you get this stuff, how should you be thinking about that? And this is why you founded Scrunch. So I think we ought to talk about how marketers should be thinking about this, but also what Scrunch actually does to make this work for the marketer.
- [14:05]
Chris Andrew
Yeah, I think like any shift, there's a huge education and awareness piece here. So if I'm, if I'm leading a marketing team and this, this is how I'm finding most of our discussions go. We start by presenting how your organization is showing up in traditional search and Then we compare that to how you're showing up across the top assistants and models. Here's an organic search result on Google, here's that same result on GPT, on Gemini, on Perplexity. And the results are different. Sometimes you don't show up at all for the same result across different assistants. And so when I think about putting my marketing leader hat on, I'm thinking about educating my team on this shift and thinking about how do I monitor what's happening at scale. You can go run these searches yourself and you should be, you should be familiarizing yourself with these technologies. But what we found as a team building technology was we came across a couple of Fortune 500 companies that had someone doing manual AI search results and putting them into Excel spreadsheets to track. How are we showing up against our top 10 competitors? How are we showing up across the five different models? How is it evolving over time? As someone who loves technology, I'm like, that's a technology problem. We can monitor that at scale, report on that, and with that foundational data set, we can start to give you the insights on how you optimize and address it. So Scrunch was founded to help you get visibility with monitoring the insights to drive the change and then drive the outcomes you'd like.
- [15:28]
Mike Linton
So one of the things when you talk to a lot of the folks in AI and studying AI is they actually don't know how the model works. In fact, we had Tom Goodwin on who said, having eye model, it's like having, you know, 1 million teenage interns working for you. They're very earnest, they produce a lot of results. They can't totally be trusted. And he also said, and a lot of people don't understand how the model is actually producing the answer. They just see the answer and they've, you know, trained the model. Tell us what's going on there.
- [16:06]
Chris Andrew
Yeah, I think, I think as a human there's a little bit of playfulness there that we kind of like. Like we don't like perfect people. Like, if someone presents themselves as completely.
- [16:15]
Mike Linton
Perfect, that's why everyone likes me.
- [16:17]
Chris Andrew
Everyone likes Mike, right? That's why I'm doing this podcast. But if you think about the mistakes that GPT and Google Gemini were making when they first came out, right, people would be like, you know, how much glue should I put on my pizza? How many rocks every day should I eat? These things were making mistakes and they're goofy. And honestly, that stuff kind of goes viral as it, as it kind of presents the vulnerability. And hopefully as Humans makes us aware that we still need to think for ourselves. Now, what that means is these models are getting better. And I think fundamentally, if I was a marketing leader at a large company, I would not bet against the models. I would not sit here and say they're not going to be good enough to answer questions for humans. We're already seeing that individuals are trusting these things. And I think where trust exists, the trust is going to grow. I think what we need to be thinking about is if that's how consumers want the information, they're willing to take some mistakes along the way. And I think you see the shift in the model starting from playful image generation to much deeper consideration like, hey, make an image for me like this. Okay, this was kind of fun. I trust it. Hey, plan a trip to Italy for me. Hey, I'm a B2B enterprise marketer. I need to evaluate the top five CRM solutions and you've got access to all the data in the world. Help me evaluate the pros and cons. I think being with the consumers as they go through that shift is something that most organizations are starting to think about.
- [17:39]
Mike Linton
Hey, but let me, let me follow on this question. If I don't know how the model actually is producing the answers, how can I adjust my kind of presence for these bots if I don't know how they work?
- [17:55]
Chris Andrew
Yeah, the first thing you can do is measure it. If you don't know how they work because they're generative, they're non deterministic, there is not an exact recipe to the results you're going to get back with that knowledge. I think the software technology problem becomes even greater. I want to monitor at scale. So what we're doing for a lot of our customers is, hey, these are the categories of things we need to be showing up for. These are the competitors we care about. These are the types of customers we care about. Go measure how we're showing up and generate thousands of responses for us and give us that visibility. That would be the first step. The second is to understand the workflow behind the scenes in terms of how these models are arriving at a decision. So we talked a little bit about web search, right? These models can't go look at every single website on the Internet when it's trying to give you a real time result. What we see in the current behavior from GPT is it's browsing, it's looking at the sites that it's thinking is most relevant. It's heavily prioritizing page snippets, page titles, it's evaluating the pages that it thinks are most relevant. It's downloading the top three to five and then it's answering the question, even simply understanding that workflow. Behind the scenes of I take a complex prompt, it turns into a simple search. The model looks at 20 websites, it downloads five. If you can understand that funnel, you can start to understand how the models behave and get insights into content gaps, metadata tweaks, page snippet structure, questions that would inform the changes you may want to make to show up more effectively.
- [19:27]
Mike Linton
And I don't know if you can. So this is a question you can take or leave without naming a company, because I know you can't name a company. Can you give us an example of how you looked at this and then went back to the company and said, here's what you need to adjust to rise up in the AI results?
- [19:48]
Chris Andrew
Yeah, that's the flow of every meeting we're having with every customer. It's literally like, here's your current search results, here's your results across the leading assistant and AI models. Are you okay with that? Sit with that question. You're then usually on the education cycle, but then it's, let's start to test some of these changes. Let's start to address some of the content gaps. Let's start to address some of the simple page snippet and title questions which are being prioritized in the web search component for AI search. So we've absolutely tested that end to end and been able to show that look, you weren't present in results for these categories over the last month. Now you're starting to appear, your content's being crawled, it's being recognized. We can see the user agent, you can actually see this, your web logs, you can see the user agents from GPT, from Meta actually browsing your site, accessing information and putting it into an answer. So it starts with a measurement of current state, ongoing querying and prompt generation. To understand what's changing, test those insights and changes to then show how you're improving or not improving against those results.
- [20:53]
Mike Linton
And so am I essentially running like thousands of a B test on things like snippets and titles and placements and everything else? Or what is that, what's going on? And then I, then I rerun the model and see what comes up. Is that. So is it brute force or is a lot more than that?
- [21:09]
Chris Andrew
Well, there's an automation component to the brute force. Right. You can start to understand what the models prioritize and how they like content to be represented. And I would start with even the simpler thing. What do these models not understand about your brand? And ask them if I was a marketer, what, what do these things not have full clarity on? Maybe they don't understand your return policy. Maybe they don't understand your international shipping very well because they're pulling from a bunch of sources and the information is conflicting. I would look at gaps in content, I would look at inaccuracies and answers and think about creating content to address those gaps. But then that AB generation is something that we're helping companies automate to understand. What are the models looking for? How can I test against that at scale? Which is why we're moving from not just monitoring but into insights and kind of measuring the outcomes.
- [21:58]
Mike Linton
Got it. Thank you. You also mentioned earlier that, you know, search, as search and consideration drops down, you know, Google will lose its stranglehold on search or searches. They might keep the stranglehold, but it won't be so big.
- [22:12]
Chris Andrew
Yep.
- [22:13]
Mike Linton
What's going to happen with the other models like Bing and everything else are like, what happens in the search space now?
- [22:19]
Chris Andrew
Yeah, I think it's going to be fascinating to watch and I think it's going to unfold very quickly. Search is powered by ads, right? I mean, has there ever been a better revenue generation model in the history.
- [22:30]
Mike Linton
Of the world than we're going to think about it? Yeah.
- [22:32]
Chris Andrew
Right. It's just like the margin's insane. It works. You can measure it. The fact that Google is putting an AI summary above their paid search results, above their lists, above their maps says it all. Like the fact that they're willing to do that means that's where it's going in. Their hand is being forced. Now Google is going to play exceptionally well in this, in this situation. They have the data, they understand the consumers. But I think what we can expect to see is we're going to see new ad formats arriving where you're going to have to pay more. Perplexity just introduced ads. They're the new player. They need money more than the incumbents. I think I heard their ad formats cost about 10 to 20x. What the traditional search providers, why you're so much further down that purchase experience and buyer funnel by the time you're making a click. So I think what we can expect to see is Bing is relevant again because GPT is using Bing to generate search results. Meta search is becoming relevant. Perplexity, a new player. How do you become an upstart in the search category? That's remarkable. They're relevant. Google's going to stay relevant. GPT is relevant. I think we'll see 5 to 10 core search providers and distribution is going to be a huge question here. The fact that Apple Intelligence has partnered with GPT means you're going to be able to leverage that model to get an answer. So I think we're going to see a real shift in kind of the search landscape. Not over the next five years, over the next six months.
- [23:56]
Mike Linton
Wow. And if I, if I look at all these search models, are they essentially just raw material in the long run for AI or are they more than that?
- [24:06]
Chris Andrew
Yeah, it's a, it's a great question. I think, I think they are essentially raw material for AI. Like, I mean, AI wants content. AI is content hungry. It's trying to make sense of this information and give the answer back to the consumer. I think we'll see some significant changes in terms of how businesses think about presenting information to the models. Right now the model's got a pretty tough task of sorting through all the interactive components of your website. All of the different video things, the imagery, simple structured content that addresses the question that the consumer is asking is going to be valued. But I do think thinking of your content as raw material for the models is a good way of thinking about it.
- [24:48]
Mike Linton
So this is going to restructure then all the raw material out there. If I'm, if I'm hearing you right, I'm to say, am I getting this right?
- [24:58]
Chris Andrew
Yeah.
- [24:59]
Mike Linton
Websites, everything you're releasing that is raw material for the model.
- [25:04]
Chris Andrew
Correct. And the models are going to keep getting better. Video is an input for the models, imagery is an input for the models. But right now we're seeing simple, structured, almost table based FAQ information is very effective in these models. They are looking for the answer. When it is well structured and simply presented, you're effective in getting returned in the results.
- [25:25]
Mike Linton
So this is going to obliterate not just raw materials, but a lot of jobs around those raw materials. Am I getting that right?
- [25:32]
Chris Andrew
Yeah, I think it's kind of fun to sit back and be like, this is going to be the societal or career based impact. I think these are tools for marketers to be more effective and engaging with their consumers. And I think if you think about things that are going to become even more relevant, it makes me think of analog. It makes me think of like, how do you be deeply personal with your customers in ways that truly still feels personal? There's a lot of AI customization tools that like, make it seem like a personalized text or email to Mike and people are seeing right through it. They're like, this was just grabbing the last thing I said on LinkedIn and putting it in a message to me. Oh, yeah, good, right. I think there's going to be an opportunity for humans to be more human, use these tools to do the automated generative tasks that have been quite manual in the past and free us up to be more creative.
- [26:22]
Mike Linton
Now, a couple of our guests have said, in this world, brands matter more than ever because if you don't have a good brand presence out there all over, the chance of the bot not finding you is stronger. And is that a right answer to branding or not? And I'm thinking more in the B2C space or even in the B2B space where, you know, you've got to get in the call or you've got to get in the evoked set. How do brands play in this game?
- [26:53]
Chris Andrew
Yeah, I mean, if you think about a customer journey, that awareness stage, if the human isn't aware of you because your brand isn't strong enough, you're not considered right. If you're buying a bike or a jacket and I don't know who you are, I don't value the brand that you've built. You're not in my consideration set. Same exact thing for AI. But going back to your original statement of your best customer is an AI bot. How do you brand for the AI bot? How do you make sure the bot is aware of your brand? And I think traditional brand building exercises are very, very important. But I would also think through the very simple takeaways, right now, these AI bots are not watching tv. Right. These AI bots are looking for content. They're trying to make sense of the world and the way that they're able to make sense of text and generate an answer. And so thinking ahead about what you can do to be brand aware for that new consumer that AI bought is, I think, a fun exercise to be running a workshop on as a team.
- [27:48]
Mike Linton
Thank you for that. So let's stay on this marketing theme. We've got an awful lot of marketers and agency folks listening to the show. What should they be doing right now? And then next year with all this stuff, Give us some concrete things.
- [28:07]
Chris Andrew
Yeah, I think there's, this is such an opportunity for marketing because there's such a desire to run AI projects. Every company is trying to be more AI.
- [28:16]
Mike Linton
I mean, if you're not doing it, you're going to get, you're going to get fired. Your board or your CEO.
- [28:21]
Chris Andrew
So public markets companies are getting rewarded just for saying AI and earning calls. Right. Whether there's anything behind or not. So, you know, in terms of what I would be doing if I was a marketer is one of the benefits of consumers already changing their behavior is you can go see these results. You can go see how your brand is being represented in these models. What consumers are learning when they're searching for you, your product, your service, your competitors, the industry that you're in, that's a project that's pretty low risk. To start a lot of these other AI projects, they need access to your data. There's problems with hallucination because you might misrepresent your brand. The reality of AI changing the customer journey is it's already changing and generating live results for the consumer. So you need to start paying attention. So the first step I would take if I was a marketing leader is I would just sit in a room with my team and start testing and seeing how we're showing up across these models compared to the organizations that we're.
- [29:16]
Mike Linton
Competing against and in the people you're talking to. Is anybody able to start tying this to actual sales or are they just tying it to like, maybe leads or like, how are they measuring the impact on this? Because eventually you're going to want to say, I want to pour a ton of money in this because it drives this kind of sales or this kind of leads or this kind of appointments. Is anybody doing that yet?
- [29:41]
Chris Andrew
Yeah, we have not seen the full tie through to dollars. And as a company, that's our objective to get there. I think it starts with monitoring. I think it starts moves to generating insights, then it starts to measure the outcomes. Right now that data is not accessible from the platforms. What you can see though is these user agents, these bots from the top models visiting your website. I would be all over your web logs with your IT team and marketing team to understand the frequency of when these things are crawling your site and what information they're accessing. Now, the volume is not going to be the same because they don't need to crawl your site every single time to get this information. Once they've got it, they're returning it. Now they're all commercially minded. These models are losing money hand over fist. So what I expect from an outcome perspective is when I get an answer via my assistant, I'm going to ultimately have a link to click to book a meeting, to buy the product, to click an ad. We're going to see new formats introduced that measurement is absolutely going to come and that's going to challenge a lot of the traditional search ad volume that we've seen over the years. So I think outcomes are coming and it's really about measuring, optimizing current results to get ready for those outcomes.
- [30:53]
Mike Linton
Because once if you actually create real ways to measure the outcomes like happened, you know, when search came out, this thing will explode for you as a marketer, for better or worse. So. Okay, interesting. I think that this gets us to our traditional last question. It's a two parter. You can take one or both, but you have to take at least one funniest story you can tell on the air or practical advice to our listeners we haven't discussed yet.
- [31:22]
Chris Andrew
Okay, I think I'm gonna go practical advice. Plenty of stories that would get get in trouble, get me in trouble here on the practical advice. It's very simple. I think it's, it's think for yourself. I think we're entering a world where critical thinking, taste and your perspective as a brand leader, a marketing leader is the most valuable asset you have. I think taste is something that people can still feel and they don't quite know how to pin pin it. Now these things that are influencing your critical thinking and taste, there's a statement that you are the people you spend time with. You're the books you've read, right? Your parents, your teachers are always the people you're spending time with. It's the books you've read. The new component in that list of things of the people in the books is I think it's going to be you're going to be the assistants, you're talking to the AI agents that are organizing information for you. If you stop thinking for yourself, I don't think you're going to be as valuable in this world. So my advice, my practical advice to business leaders, marketing leaders, is to pause. Think for yourself. Think about the impact to you as a consumer and think about the impact to you as a business leader.
- [32:27]
Mike Linton
I think that is a great way to end the show. Think for yourself but recognize what's coming down the road as you're thinking. So thank you Chris and thanks to everyone for listening to CMO Confidential and thanks to props cl, our sponsor. Look for more shows on the I Hear Everything network, Spotify, Apple and YouTube which include marketing the battle between believers and non believers parts 1, 2 and 3. The rise and fall of peloton as seen through the eyes of cltv Using AI for anticipation versus Reaction and its AI an extinction event for agencies. Hey all you marketers stay safe out there. This is Mike Linton signing off for CMO Confidential.