You Will Lose Money If You Don’t Use AI This Way

CMOs are jumping all over the AI train, but most are doing it wrong. Let's break down the biggest mistakes and how the smartest CMOs are doing it differently. In this video, I'm going to walk you through the most common AI mistakes CMOs are making and how the smart ones are leveraging AI for actual business growth, operational efficiency, and better decision-making. The first mistake I see is launching vanity AI projects, showing off chatbots or AI demos without clear metrics, goals, or business impact. At the end of the day, if you can't show that you're moving metrics in the right direction, that you're acquiring more customers, that your conversion rates are going up, that your lead flow's going up, or maybe your cost per acquisition is going down, none of it matters. If you show off a chatbot that's generating a bunch of conversations that don't do anything, who cares? If you are creating a bunch of AI slop, maybe you're creating a lot more content, that's not the actual metric that matters. Are you actually getting more reach? Are you getting more engagement from the content that you're creating? And is it engaging the right people? 'Cause if you create wide TAM content and millions of people see it, but it doesn't drive any more customers, why does it matter? CMOs need to focus on not the high vanity stuff that drives views, but actually on driving business results. When you think about high leverage trackable AI projects, in contrast, smart CMOs start with single P&L linked projects like reducing MQL costs by 15%, then tracking results in terms of CAC and payback periods. What I'm more interested in seeing is, okay, well, if we're going to be producing more webinar content, for example, okay, what content is going to resonate? How should my headline and thumbnail be for my ads to drive people to my webinar? And then how much am I acquiring these webinar signups for? And is that cost going down over time? And are we actively looking at the number of MQLs that we're driving per webinar? That matters a lot more than just creating wide TAM content and saying, oh, we should just do YouTube Shorts and just try to drive a bunch of views. That isn't what moves the needle. The second mistake is actually in content obsession. So content is important, but AI shouldn't be used as a content mill. So earlier in the last one, I talked about people generating a bunch of slop. I see people sharing these end-to-end workflows where they're pulling from the highest performing YouTube videos out there that are trending, and then it's making them into these faceless, YouTube long-form videos or short-form videos, but it's just creating a bunch of junk on the internet that people don't really need to see. And that means it's an opportunity for you as a leader, as a CMO to say, okay, well, if people are going to be creating a lot more content, how do I zig where people are zagging? So the smart play here is actually layering on decision AI. So you can have budget shifting, routing, scoring on top of content creation. So instead of using this end-to-end workflow for producing content slop, you can use it more smartly. And what I mean by that is using it to help you create better packaging for YouTube, better thumbnail ideation, better headline ideation. Or you can use it for having a note-taker join your calls. It looks for compelling ideas that are most likely to go viral and it scores those ideas on a scale of 1 to 10. It gives you 3 different headlines and then it gives you the body copy. You still need to do work as a human. Don't try to automate everything. And a way you might want to think about this is before you automate, do it on a manual level and then delete things that don't make sense. And then once you think you really have it down, then you can automate it. I think too many people are trying to automate these things and content is one of maybe the last things that you should automate if you want to have a brand that actually stands out. Now, the third mistake I see is not setting any guardrails within your company. AI can lead to brand and legal risks that actually hurt your reputation. And so, for example, I've seen some people make ads using Google's V0.3, which is amazing. It's an amazing text-to-video model and you can have stormtroopers or you can have Darth Vader reading your ads out, but that's going to get you into trouble if you don't have any governance around how you go about using AI. So you've got to create AI governance policies or contribute to AI governance policies in your organization. So the smart CMOs, they actually create and set clear policies with data classes. They review tiers. And human-in-the-loop thresholds to make sure that things don't go off the rails here. Because AI is super helpful, but you can't just let it do its thing right now. Otherwise, it's going to cause you a lot of headaches. So the 4th mistake I see is when CMOs ignore the cost of model drift and upkeeping models. And what I mean by that is you have to make sure that you're constantly retraining the data that you have to make sure that you actually have model accuracy here. You're going to have the models that will hallucinate. You're going to have to have a human in the loop. You can't just let this work on its own. You're going to have to continuously feed it new data because oftentimes it's one, it's going to hallucinate, but 2, These LLMs have limited context windows and they're getting stronger over time. ChatGPT does a good job with remembering things, but even then it still forgets a lot of things where you have to feed it information. Each chat is a new chat and oftentimes it's missing a lot of context, right? So your job as a CMO is to make sure that whatever important flows that you're running, that you're constantly making sure that you're feeding it new data, that you're also updating the newest models. For example, if you're using Lindy, that you're using the latest models in each part of it and make sure that drift isn't happening and make sure that there's alerts happening if models start to drift. I'll give you an example here. When we have our coaching bot, sometimes it'll start to call come out with lesser and lesser outputs and it'll start to forget about people and it'll miss people and it'll start to make mistakes. And so you want to make sure that you're constantly calling out the mistakes, you're giving it feedback just like you would with a human being. Because again, these are human spirits that you have to stay on top of. So problem number 5 is data hygiene. So AI needs clean data to perform and AI doesn't automatically fix poor data hygiene. If your data is messy, the outputs are guaranteed to be garbage. And so you got to be working with your data team on this, your data scientists on this to make sure that the data that you have is, is clean and make sure that AI is even helping you clean up this data too. So smart CMOs, They understand that clean data and unifying the data layer before buying is going to help you do a better job with AI, right? You want to help AI help you, and that's where this comes in. And number 6 is where CMOs that treat AI only as a cost saver. This is the CMO that cuts headcount but misses the growth upsides. If you're hiring people that are AI-enabled and they're 10x what they were before, you're going to gain way more leverage out of that person than someone who isn't AI-enabled. Now, someone who isn't AI-enabled, you're probably going to have to let them go at some point, especially if they're unwilling. But you don't want to look at it that way. I mean, throughout In history, we've always looked for more leverage. It's labor leverage. You have capital leverage, you have code leverage, you have media leverage right now. If labor is 10 times or 100 times stronger, you want those people on your team because it's still a lot of leverage. And so instead, you should continually reinvest your saved budget into innovation loops. That could mean hiring more people or reinvesting into new agentic flows or new agents that can help you do work at a higher level, more work at a higher level at lower cost. The 7th mistake I see from CMOs is delegating all AI decisions to a specific department like IT. So recently we're just talking to a company that does about $30 billion in revenue and they have a lot of employees. And they just said that IT is bottlenecking a lot of these AI decisions 'cause they're not following a lot of this stuff, right? And the teams that are trying to push these new AI initiatives, they're getting put on the back burner 'cause AI is bottlenecking these, right? And so what you don't wanna do is you don't wanna delegate all AI decisions. You need to have champions within your marketing department to push these things, right? So the smart play here is involving data, brand, legal, marketing, RevOps to ensure that AI is in line with business goals and having champions push that forward. Here's the problem with most marketing agencies. They don't understand what is happening With the world of AI, with my agency Single Grain, we think about agentic workflows. We think about creating agents for the future. When we think about SEO, we think about creating different products such as internal linking products or content consolidation products. And on the other side of things, we think about building products such as Care. We just think about making marketing easier overall. If you want to stay at the cutting edge of marketing and at the same time leverage the power of AI, go to singlegrain.com and we'll see you on the other side. This brings me to my last point, which is the. Which is not championing AI within your organization. So in the last example I gave you, we just talked about having to push it and spread this positive virus throughout your organization, right? Because if you don't do it, you can have leaders trumpet it from the top, but you need people within your organization bubbling it up and pushing this movement forward. So if you're not doing that, you're not going to drive AI fluency. And what that means is you're going to have a bunch of people who are super AI enabled and you're going to have a bunch of mediocre people. If you think of your organization as a high-performance sports team, that's not a fun place to work. You have to make your brand appear as AI forward as possible. You have to You have to hire AI-forward people. You have to spend your own time investing into AI, and you have to think about when you're interviewing these people, hey, what are you doing with AI right now that's innovative? Okay, can you gimme some examples there? And if they can't do it, unfortunately you can't hire them. And then within your organization, you have to create an AI fluency program where you're driving people on, hey, here are the tables, here are the standards that we're expecting. Here's what we're gonna expect in 6 months, 12 months or so, 18 months or so. And you're gonna get there as an organization. And you should tell everyone, if you're willing to get there, I'm gonna help you. If you're unwilling, unfortunately this is not gonna be the place for you for the long term. So there you have it, the biggest mistakes CMOs are making with AI. And by implementing AI the right way, you're gonna to save yourself a lot of time and set your organization up for long-term success. If you want to deep dive deeper into how I'm applying AI into my business, you want to check out my live talk at the YPO Global Marketing Summit, where I share my best AI business hacks for scaling quickly. Check it out.