Why MCPs Are Secret Revenue Machines

Most people don't realize how many ways MCPs can actually drive revenue. You don't just want to build because you can. Ultimately, what you build should bring in direct revenue. With MCPs, you can basically create these mini employees that never sleep, pulling insights, running experiments, and surfacing moneymaking opportunities you'd never see otherwise. Today, I'm going to dive into the biggest revenue use cases for MCPs I've seen actually get implemented into businesses. Let's get into it. So just a quick refresher for everyone. MCPs are model context protocols. The way you want to look at this is it's very similar to APIs. It's just having your applications be able to talk to AI or these LLMs. 98% of people are saying they're struggling to see a return on investment from AI. But when you look at the 2% of people that actually are driving a return on investment from AI, of that 2%, 61% drive actual revenue business results from MCPs, model context protocols. One of my favorite use cases is what I call the CRO orchestrator. CRO stands for conversion rate optimization. I'm just trying to simply figure out how I can increase my conversion rates for my website. So let's say 100 people visit my website and 1% people convert. Well, if I can double my conversion rate, then that's 2. Well, then if all other things are held equal, then that means I'm in effect doubling, right? A lot of people don't think about optimizing their conversion rates on their website. And also a lot of people aren't conversion rate experts. And so what you can do is you can use something like a, like a Claude code, and then it can hook in with my Google Analytics or my Google Search Console and look at the data that we have and I'll look at what pages are maybe worth going after first that are driving conversions, that get a lot of visitors, and I'll look at what the low-hanging fruits are. And then what I'll do, it will actually point out what we should be optimizing for and what the impact, confidence, ease, the priority should be for each of these tests. Right. And then I can send that to my developers and we can run these experiments. And we've actually done that on my website. We have seen a lift of maybe 20%. Maybe we change a headline on, on, on a page or maybe we add a new modal, we add a new offer. These things all matter. Human psychology doesn't change, and so if you're strapped for time and you need ideas here, what we're ultimately looking for from these MCPs is the velocity of ideas to increase so we can run these ideas, we can fail faster, and so the company is going to grow faster because the company is the sum of the problems that we ultimately solve. Next up, we've got sales intelligence, and you can connect your MCP to a sales intelligence product like Gong that listens in to our all of our calls with our clients. And what we can see is we can see the questions or objections that keep coming up on calls. I can just query whatever my question. Are, and I can look at interesting trends that it's spotting with maybe deals that we've won, deals that we've lost, and then we can figure out how to continually adjust our offering and what our products are, what our services are. We're able to what? Move faster. Beyond call data, you can plug in MCPs with your CRMs. So your CRMs are your customer relationship management tools, and you can be using HubSpot, you can be using Salesforce. This is your, your source of truth when it comes to a lot of your data, when it comes to closing sales or customers, for example. Now what you can do is you can log into ChatGPT, for example, And you can ask HubSpot in the last six months or so, what are the best services or products that people are picking up, and what services aren't doing as well? And beyond that, how long is it taking us to to close a deal? And what you can do, by the way, when you leverage OpenAI, you can leverage deep research. Deep research means it's going to do a deep dive of the data that you have based on the timeline that you specify. Imagine being able to query whatever it is that you're looking for, get a deep report, show this report to your leadership team. You don't have to wait for a data analyst anymore to get these things done. And then eventually, what's going to happen is you're going to have an AI that's just doing these things on a its own and they're surfacing these data points. And maybe you as an executive, you're just getting text messages on key trends that are happening and then you can make decisions based on those trends, right? Another killer use case here is follow-ups. And so if we just look at the last MCP use case, you can simply connect your MCP with your Google Calendar, also your Google Mail as well, and you can hook it in with HubSpot and you can just say, hey, based on all the meetings that we've had in the last 12 to 24 months or so, maybe these are deals that we've lost. Who should we be following up with? And can you draft the email for me. Now you have the agent doing the work for you, and you can even say, hey, my ICP, my ideal client profile, is companies that are doing more than $25 million a year, and I want you to search based on that criteria. Do deep research. Sometimes this thing might take 20 to 30 minutes to do the research, but it's looking at the last 24 months of data that you have. And how much more money is that going to make you? A lot. Everyone's talking about AI right now, but people are struggling to prove the business ROI of AI. Now, with Single Grain, which is my ad agency, that is not an issue because we try to stay at the forefront of AI. We leverage AI not only to make our team more efficient and faster, but to make your business more efficient. And the things that we do around marketing, it's still our bread and butter. But when we combine that with our AI prowess, that's what allows our clients to grow a lot faster. So if you're interested in learning more, if you're interested in acquiring customers faster, go to singlegrain.com. Again, singlegrain.com, and we'll see you on the other side. So now let's look a little outward. I want you to take a look at hooking your MCP in with ad libraries and tracking what competitors are doing, what they're spending on Meta or Google. So we can call this competitive ad intelligence. And what we do is we hook it in with browser-based, or you can hook it in with other tools that can, that can scour the internet or API-ify, right? And the cool thing is I can say, okay, my competitors are Agency A, Agency B, Agency C. I want you to look at how much they're spending per month, what their strategy is. And it actually listened to me. It showed me, it said, okay, this competitor is spending X amount of dollars and they're only targeting local keywords that are high intent. Another one is targeting just the most expensive digital marketing agency keywords to go after, right? And so I'm getting a sense for how they're running their strategy, and that might help give me some ideas on how I might want to run my strategy. So now I might not copy them, I just look for inspiration to come up with my own ideas. Now once you have this data, you can take it even further. You can connect your MCP with Meta or Google Ads accounts and see what's performing, and you can have this automatically turn on or shut down whichever campaigns aren't working. And the cool thing too is it can look at what's working and it can hook in with the Figma MCP so Figma help you design different types of ads or different types of creatives, and then you can run them using this MCP connector. Now beyond ads, SEO is another huge play. I love SEO. Obviously my background's in SEO and I think about, okay, there's a WordPress MCP. WordPress is the CMS, the content management system that we use, and we can have it scan the Ahrefs MCP. So that's an SEO tool. It can also look at our analytics, our Google Search Console. And then what it can do is it can take a look at our blog. I can publish our writer guidelines. It can start to publish content for me that might be 80 to 90% of the way there, and then we have a human in the loop review it, and then we are now able to do what? We're now able to do SEO at scale. And SEO still isn't dead. I talk about like traditional SEO is still very much in play. You want to leverage this for 10x leverage. This is one of those plays. From there, you can consolidate all your reporting platforms into one view. And so like I mentioned earlier, you have your analytics, you have your ad accounts. All these things can be pulled into one view. And what you can do is you can create a sub-agent. You can use like Claude Code, for example, and you can say, okay, I want you to make an agent for me that consolidates, gives me a monthly report of all of my top channels. What are all the top insights? What are all the top growth opportunities? What are the things that we're missing out on? And it, it can just do a summary for you every single month or so, and it consolidates it for you. And you can continue to improve the sub-agent over time cuz the sub-agent is just simply a prompt. Okay. Finally, last but not least, think about content leverage. And so I love connecting different MCPs to look at my pillar piece of content, and then I wanted to come up with different types of content for different types of platforms. And so a YouTube video that might be, you know, 15 minutes or so could then be repurposed for X, could then be repurposed for LinkedIn or like a short-form piece of content. And so I have a content repurposer sub-agent that will pull from, let's say, Apify, for example, or Browserbase, and it'll look at, you know, who I perceive as my top competitors on YouTube. It'll give me ideas on what's performing well for them, and then it'll also help me repurpose content that I might plug into it, and then from there it's going to generate new ideas for me, new formats to mirror what's already working or what I want to repurpose. And it ultimately turns one idea into dozens of assets. Now, the crazy part here is that we're still early. Over time, you're going to start stacking each of these and then revenue becomes a systems game for you. You're going to have a bunch of these sub-agents open and you can even just query whatever data that you want directly from OpenAI to whatever connectors that are, that are happening. And we're going to see your favorite apps connecting in with these systems. You can have OpenAI make designs for you based on whatever it is that you upload. Look, if you watch this video all the way through, let me know what your thoughts are in the comments and drop any ideas on other heavy use cases. I might just build them. I'll see you in this next video here on how to scale even faster with AI.