Podcast: Play in new window | Download
AI (Artificial Intelligence) is quickly becoming a major force in the content marketing landscape. Although it holds exciting potential, its misuse remains prevalent. Jeff Coyle from MarketMuse is here to help us explore how AI can be best integrated into our content workflow.
Rich: My guest today is the co-founder and chief strategy officer for MarketMuse. He is an AI content marketing expert with more than 21 years of experience in the search industry. He’s focused on helping content marketers, search engine marketers, agencies, and e-commerce managers, build topical authority, improve content quality, and turn semantic research into actionable insights.
Prior to starting MarketMuse back in 2015, he was a marketing consultant in Atlanta, and led the traffic search and engagement team for seven years at TechTarget, a leader in the B2B technology publishing and lead generation.
He earned a bachelor’s in computer science from Georgia Institute Technology. He frequently speaks at content marketing conferences, including Content Tech, Marketing AI Conference, Content Marketing World, LavaCon, Content Marketing Conference, and more. He’s been featured in Search Engine Journal, Marketing AI Institute, State of Digital Publishing, SimilarWeb Chartbeat, Content Science, Forbes, and more.
Today we’re going to be digging into how AI can help you with your digital marketing with Jeff Coyle. Jeff, welcome to the show.
Jeff: Thanks so much. What a great intro. I appreciate it.
Rich: It’s exciting. It’s good stuff. I love it. And MarketMuse, so that’s where you are now. It’s been in existence from 2015, and a lot of agencies use it to help them with their SEO. We use it here at flyte. I have rewritten pages using MarketMuse after Andy Crestodina, our mutual friend, turned me onto it. All that being said, I’m wondering how much more interest and inquiries there been for you ever since ChatGPT exploded a few months ago?
Jeff: It’s wild. Everybody wants to know how to use ChatGPT more than to create a poof piece of content at the end of the funnel. They know that there’s more to it. And what we’re really seeing is a lot of people trying to make a screwdriver be a hammer with a large language model. Because it has one use case, right? And it shouldn’t be, for example, used to do math complex math problems. There are artificial intelligence platforms and significant investment in that. It shouldn’t be thought of as something that’s going to give you expert level content. It’s a general, large, language model.
So a lot of people are coming to us asking, “Is this something we can use as part of workflows?” And it absolutely can be used as part of workflows. Just as a look back three years ago, we launched a language model technology into MarketMuse. It actually would create a topic expert model out of the general language model, and we built it all ourselves. We released it two years ago and it was not successful. People weren’t ready for it and it was too expensive. It created beautiful drafts of content, but you had to pay a hundred dollars for them because at the time the technology was slow. It took days to build, and it would cost $20-$30 to build one. So the economics of that didn’t make sense.
But now when it can cost basically a fraction of a fraction of a rounding error of a penny to do a generation out of a large language model that’s 80 times larger without any fine tuning. And I’m happy to get into what these words are, but basically imagine you could take a language and then tune it for a particular topic or in particular to make it have more expertise. You can train, retrain, and retrain until it might get a little smarter. But at the time it was too expensive. That was the sole barrier, too slow, too expensive. Now it’s fast and you’re going to start to see point solutions for specific workflows.
But people come to us all the time now and it’s saying, “How can I use this? How can I use this?” Because it feels so new, and it feels so magical. And it’s a really interesting moment in content right now because a misuse of it is the majority of use. And it’s almost frightening to me. But the folks who are really integrating it and understanding how to use it in that workflow and that it’s not magical end state, it’s actually most useful when it’s part of the same content journey that we’ve been evangelizing one take anyway. Use it to support research, to support planning and prioritization, to support content briefing. To give your writers the ability to potentially do things that they wouldn’t have done as part of writing, editing, publishing. And those use cases are far more valuable right now than getting it to create content that then can go into an editing cycle.
Rich: One of the things that I want to ask you about, you mentioned a few minutes ago, is honing these machines down, kind of going deeper in. So it sounds ChatGPT and its ilk are generalists at this phase of edge of pollution. They know, I wouldn’t say a little about a lot, they know a lot about a lot. But at the same time, from what I’m hearing, that the future of AI could be the future of any member of our team where we hone their expertise, they start understanding marketing, then SEO, then link building, and they keep on niching down and becoming more of an expert.
Are you saying that the next evolution, and perhaps it’s already happening, of AI will be creating these topic specialists? So we might use one AI for marketing a motorcycle company, and another one for marketing a dog grooming salon?
Jeff: I would hope that’s the evolution. I would hope that, especially when you look at the product management team at Microsoft, for example, which is one of the best product management teams there is. You open AI, hopefully they’re not on this call, but they’ve shown that product management isn’t where their focus is. And giving everyone access to something that this democratizing and not having clear plans, I think Microsoft is going to, illustrate that they’re going to have a little bit more foresight.
I would imagine that, for example, the Azure team will come out with a fine-tuning solution for businesses, where you could take advanced language models and train them or tune them for specific tasks for specific needs. Maybe even training them on things like post-sale documentation, technical documentation, style guides with brand voice content, governance practices. I would imagine that you’re going to start to see those types of things roll out that are much less retail solutions and getting into B2B, and Microsoft’s the perfect company to do that. So I would expect that. Fine tuning and post tuning is expensive still. It is. And that will come down and become less rigorous to manage.
So those are things that I would expect there to be point solutions of all types, shapes, and sizes. And they’re not going to only come out of language models, you’re going to potentially have other items. But you think about a language model, it has to have sources of content to be able to make… all it’s doing is predicting. It’s predicting the next most likely word that it should write and then it’s progressing. Some of them have short term memory, midterm memory, longer term implementations.
But for example, it can’t predict something that it can’t predict, right? So if you’re asking it to do things that it can’t possibly do, it will make it sound convincing. And what you’re seeing right now is people trying to ask a language model to do something that it wasn’t ever meant to do. And then they’re saying it’s broken, and it’s a little bit odd. You’re seeing all the noise in the system of people going, “Look at this thing, it’s broken.” And you can make language models. Chatbot implementations are the equivalent of prompt-based ones. You can make them steer very wrong. And I certainly can do that in minutes because I’ve been working with these things for so long.
But I would just say stay out of the noise and start thinking. Take a personal content process inventory. Your inventory, your processes, your teams. All the manual steps, all the tools you rely on. Write them all down, get ugly with the list of manual steps. Be real and start to think which of these tasks should I be looking for a solution for? And then get to the point where you do it at the KPI level. What are the things you think can’t be predicted or can’t be optimized? They probably can be.
And so one example of that is marketing. At MarketMuse we’re looking to enable teams to predict the content they should create or update right now, that’s going to have the highest chance of succeeding. Almost like a content batting average or an efficiency rate. It’s not just about making you faster, it’s about we’ve only got the capacity no matter what to launch 20 articles this week. And you already know from history that only two of your articles out of every 20, one in 10, is successful. That’s tremendously inefficient. I don’t care how you build that content. You’re one in ten, that’s going to be your number. So what are your options to get 10 winners? You can write a hundred articles. Or you can try to make 30 amazing ones, batting 50%. And that’s the journey I take teams on is thinking differently about what it is that you’re going to get out of the bottom of that pool and that process.
And then you get into content updates. How do you update content? How do you decide what to update? When do you update it? Are you confident that it’s successful? Those are another process. Another is repurposing. That’s a huge forgotten use case of artificial intelligence and language models. It’s to say, I spent all this time on this massive piece. It’s great. It’s doing really well. What are the different manifestations of this piece? And artificial intelligence helped me to maybe build a landing page for each target industry that is inspired by my 5,000-word opus.
Rich: That is a brilliant idea. I’m totally stealing that. Because we often talk about account-based marketing and taking 90% of your content and then just tailoring it with 10% for doctors, for lawyers, for brick layers, whatever. So that, that’s actually great. And very often, we’ve talked about should we be doing full transcripts or should we be doing show notes on this podcast. And so far, I haven’t found the right AI tool to summarize a back-and-forth interview into a coherent show note. So I’ve stayed away from that.
We can talk about that. And I do want to get into the mechanics. Like I know that people who are listening are like, okay, but how does AI, how does marketing tools like MarketMuse help my SEO? But I do want to ask you one more crystal ball kind of question. And you talk about the idea of tuning, and so that we might use different flavors of AI for specific industries. My question is, at what point, since basically they’re predicting things based on things that have been written in the past, at what point does this become an echo chamber, and at what point is there a sea of sameness in AI created content? And if that’s going to happen by default, how do we avoid that?
Jeff: Yeah. I mean there is. If misused, right? Not only can it be a sea of sameness, but it can be a sea of wrongness. And this is something that I was just recently speaking about. Because if there is no clear source of truth on a declarative, that’s one thing, right? It’s this team won the NHL Stanley Cup last year. If you say that a different team won it, that’s wrong bluntly, right? But if it’s a piece of advice, if it’s something that can’t be wrong, that’s when the sameness can appear, right? It’s what are the most important things you could prepare for a Thanksgiving dinner? And all of the top 10 lists are the same, right? And that’s the sameness concept. If that isn’t insightful, if it’s not providing more value, the thing that you produce, you then create more of the predictions, theoretically.
Obviously, this is at massive scale. More of the predictions will be your conflictive information, and less will be anything differentiated or inspired. So the same, it can propagate, but it can also go way off track if we held that scale. So there can be incorrect stuff propagated, which can be a downward spiral. There can also be derivative. It’s not that it’s being plagiarized or copied, it’s that pool of content is contributing to the predictability of the next of content if executed that well, which is why I believe, and if you want to go in the crystal ball, it’s the subject matter expert’s the most important person in the world. In every single individual topic. Because they’re the only ones that can validate an answer being both right and advisable if it is not a right or wrong.
So the subject matter expert editor, someone who understands content operations, content management, and the topic is the diamond right now that nobody knows and nobody’s giving the credit. If you’ve got editorial strategy, marketing stops on top of that, you’re even more deadly. If you can learn how to do prompt engineering at the same time. I mean, these are ways that’s going to be stratified once this irrational, immature playing with a toy becomes boring and people realize it’s not actually working the way that they think it’s working. Those people are going to be the last ones in the room going, “See, I told you so.” Now let’s move on and make this work for our business because we can’t afford to publish a hundred articles on our site and watch our site crash and burn because we’re an actual business.
If you were the bestseltzerwater.info, and you get clicks to your Amazon link and you put out a million pages on seltzer and then the site crashes, what you do? You go buy myseltzerinfo.org and you try it again. And you fail, and you fail. But if you’re doing work as an agency, if you’re working for a real company that you don’t want to disappear, you can’t take those risks. You can’t gamble with the B2B technology company’s website that you’re going to throw out a hundred thousand pages on it.
Now, these are exhibitions in futility at the business level for someone to go, “Oh, I published a billion pages, and I made a hundred thousand dollars.” First of all, no you didn’t. And it’s going to crash tomorrow, or at some tomorrow in the future. And so you’ve got to be the one standing there going, let’s get these into practical business usage. Let’s be the thought leader of how to make this work for us. We just cut our prioritization time. We just cut five meetings that were brainstorming. We just cut our editing cycles from four to two. That’s what’s really going to be meaningful in the end and for content that’s going to go on the web and represent us. There are other use cases, temporal, social, email, those types of things. But if it’s going to represent our business and be a source of truth on the internet, this isn’t a use case.
Rich: All right. So let’s talk about this. Let’s bring it down to earth and let’s talk a little bit about how businesses can use AI specifically around SEO issues. And so it does bring up an interesting point though, because Google has more or less admitted that it sees and views ChatGPT and similar products as a threat. Which is why they launched Bard before they probably should have. Do we need to worry if they see it as a threat, do we need to worry that our AI generated content creation is going to hurt our search rankings?
Jeff: Yes, most certainly.
Rich: And I’m sure you have a solution to that then.
Jeff: Yeah, of course. First of all, it’s natural language generated that ChatGPT is by no means the first entree into generated content. Generated content has existed for more than 15 years in various shapes and sizes. There are beautiful implementations of natural language generation by professional publishers as early as like 2015, 2016, 2017, 2018. Some of the technology that we built at MarketMuse was inspired by some of those. Washington Post, for example, had their own internal system called Heliograf, that was very inspiring for me when I was thinking about ways to do this. It can also be the ways that people think about this differently.
So to answer your question, generated content, if it provides user value and doesn’t fall into the bear traps – which I’ll mention – can contribute to success in organic search. Absolutely. But the dangers are there, so you must manage the dangers if you take on the generation. And those dangers are catastrophic for SEO, potentially. They could also be catastrophic for your business’s public relations, PR. It could also be catastrophic for users, right? For one reason or another.
And so if I had to classify these things. If Google were to, for example, which they do evaluate topical authority at the topic site section and topic page level – mostly topic site level. So saying, how authoritative is this site on this topic effectively, or site section on this topic. And collection of your content contains bad advice, quality content, quality generated content, or wrong content on that topic, it’s going to negatively impact you, grammatically negatively impact you.
So if you’re also plagiarizing, if you are spinning content from another source and you’re just rephrasing it, it can have that same devastating impact. If you are using this and morphing it into something that’s magical and it’s truly an assistive technology, that’s the exact same level of assistive technology that one might get out of Google’s Smart Compose, which is an NLG solution, or Grammarly.
Rich: Or to bring it into the real world, using a table saw instead of a hand saw. If you use a tool, it’s not necessarily bad, but there’s got to be the human element at the beginning and the end of it. There’s got to be the prompt. And then I think, I believe there needs to be that human empathy that goes into the end product. So you can use this to help you, to advise you, to support you, but you cannot just press ‘auto run’ and expect to get SEO content that’s going to shriek to the top of page one.
Jeff: The gotcha is the same gotcha… I’ll tell you about a lot of noise in the system. It’s the people don’t care if their case study dies because they’ll just do another thing, and you can watch out for that. “I did it this way once and it worked”, doesn’t mean that’s a sound strategy for longevity.
The example I like to use is nonsense word SEO. For a long period of time, people would champion themselves. Typically these are people who like to champion themselves for doing crazy tactics to get a nonsense word to rank. And they’d be like, “I can make myself number one using these tactics. So thus, those are good tactics.” It doesn’t work that way. In that sense, the first application of machine learning has been applied to Google. Does it work that way? Because that word means nothing. It’s not on any sort of semantic graph. It’s not a topic model. It’s not related to the knowledge graph. It isn’t susceptible to any of the learning of the history. Those are the types of examples that you’re seeing.
So I published a million articles and got this site to spike. Okay, first of all, you didn’t predict exactly what would happen. If you could, okay, good for you, but you saw a spike. That exception isn’t the rule. And that exception isn’t sound business strategy. And those are the things that I’ll just say are critical at the time when you have an emerging technology. The same type of professional years ago was touting keyword density, keyword stuffing.
It’s not being the top three competitors doing what they do, but a little bit better. You got to be thinking those tactics have a shelf life, as did spinning in its day. They’re certainly mad libbing, which you are templating where you’re taking a template like one imagines an a ama lib and you’re plugging in a different word so you can make lots of pages. This was done a lot in local SEO. “Hey, when you’re in Atlanta, you’re 1.2 miles from the aquarium…” and going through that type of thing. It’s still in practice today in the macro-SEO community. Some sites get away with it, some sites don’t. If you’re willing to risk not adding value with the content, good on you, but it’s a very harsh lesson to learn. It still amazes me when you look at the top 100 to 200 traffic sites, how many black cat techniques are in place. Don’t emulate your master, I’d say, because they can get away with it for many reasons and you can’t. I’m sorry.
Rich: Luckily, the listeners of this show have all passed an ethical background check, so we’re going to just talk about using these tools ethically.
Like I mentioned, I’ve used MarketMuse and it’s helped me rewrite some content on my own website that I’ve definitely seen move up. So I want to talk about two possible use case studies. One being creating new content, and one being improving a page or blog post that already exists. So whether we talk about MarketMuse specifically or tools like this, if we are looking to create a new piece of content, what are some of the ways that AI can assist us in creating that content?
Jeff: Yeah, so a brand-new piece of content, the first question I ask is, how do you make the decision to make this page? So you get into the ideation. And I always like to say every page needs a ‘why’. And so you’ve got to have that crystal clear. Not every page is something that you make because it’s going to be an organic graphic success, or because it’s going to rank X for Y. And those types of strategy fail in the long run.
And I’ll give you an example. I’ve worked with publishers and e-commerce companies where they have a search volume minimum, and that has to match the title tag or else they won’t even write the article. Over time, those companies get their trees chopped down by better content strategists. So you have to have your ‘why’, and it’s got to be crystal clear. This is because we saw data in ourselves from within, or from outside, or from another data source. We are confident that this is a quick win, or it is part of a mid to long term strategy or foundation building. All of that analysis can be assisted by artificial intelligence.
And in our case, what we do is we read your entire site. We process every page, and we extract all the entities and topics that you cover. We evaluate how well you’ve covered every one of those topics. Sometimes millions on larger sites. And we show you where you have gaps in competitive advantage and deficiencies, show you what that authority is, show you the pages that are powerful. Not just powerful because they have links, powerful because they’re well-written, they’re differentiated, they’re about a lot of things. And that can start to coach you on some ways to go.
Why would you write an article? It’s a quick win. I’m ranking mediocre with another page that’s not targeting the same intent. It’s a semantically related word that I haven’t focused on, but I own a couple other related ones. Cluster needs expansion. So you’ve got to have the ‘why’, and you also have to have the expectation before you even start to hit the first key and start to even think about it. That’s stage one, picking the right I number of items, and then understanding how many items need to be written in order to make an impact on whatever that goal is.
Rich: When you’re talking about a number of items, are you talking about basically if this is a new arena for us entirely, like if we decided I’m going to start talking about AI marketing and I want to become a go-to expert for this, that you’re saying to me it’s not enough to write one article on AI marketing. Like I need to understand that there’s a topic cluster here, and then the AI will help me maybe develop the number of posts. Not that there’s a magic number, but that it’ll help me develop a number of posts on a specific subject. Including what the keywords might be what the titles and headers and all that meta descriptions might be on that page. That’s what we’re looking to accomplish.
Jeff: Yeah, we’re looking to accomplish a predictive and personalized recommendation for relatively how much content is going to be needed. So we know if this is an easy, quick win, one page and done, or a longer-term strategy. So we know how to weave it into what we currently have when we write it, so that it’s not an orphan sitting out there that’s not interwoven into the site.
And like you mentioned, what are those other concepts that could be a part of it? And I’m not talking about necessarily keyword variance or intent focused variance. That is also part of it. So if you have a word, there’s many ways people use the word, and that intent list can be instructive as well. And so you might have seen there’s B2B content playbooks that are like, you need early-stage awareness. What is you need middle of a funnel, you need comparison, you need this versus that. You need, okay, you got to go a little bit level deeper than that to truly have success in most spaces. And that’s what AI can really provide you, those inspiring items. But then being able to set expectations that is going to be a big lift or less of a lift for every word is really critical.
You can also learn from it to say, wow, we’re more about seltzer water and we’re less about coffee. So if we want to own coffee, we’ve got to do a lot more. But we are at least about some types of beverages. So are there any bridges that we can build? I always love, the best beer for drinking at the beach, isn’t the way you get from beer to the beach. You see that a lot when people want to do bridges, right? You have to actually tell the story that you’re an expert on the beach, tell the story that you’re an expert on beer, and then you got to start doing kind of six degrees of separation until you get to somewhere that actually connects the dots that you are the expert in those two things. Because you can’t truly be the expert on kitty cats and the beach just by writing that one article. There’s a great example there, right?
Even if you wrote a 10,000-word article on the brand-new iPhone, and it was the most beautiful thing ever written. If you’ve never written a review, if it’s not going to do well. You go throw that same article on SceneNet, and it will. Why? That’s what I’m trying to bring is that ‘why’. And a lot of people go “links?” and no, trust me. It’s not just links. There’s the history of articles on the topic. There’s review content type. A lot goes into it.
And so that’s really where we provide a unique advantage. And that’s what I’m trying to bring. And the cool thing is, it works even if you don’t have a site. What?! You have no authority. You’re starting from scratch. Where do you get inspiration? Externally? You have to profile a competitive cohort that have the same intent, that have differing intent. Publish and see that last year this blog published 140 articles about this topic, and related ones acquired links at this velocity. And then they have this much traffic. So you can do competitive cohort analysis that can give you the same level of prediction, even if you don’t have any pages on your site.
Rich: Could I then take that information and maybe pitch those people on a guest blog post that would then link back to my website as a way of maybe bridging that gap?
Jeff: Yes. So ghost post hunting… that’s actually extremely advanced use case you called out. So I’m going to explain it to folks. So there’s two concepts that relate to internal and external linking that are key. And inverting those concepts, and I’m going to do this less abstract. I’ll give you a real example, this is one way to support outreach. So if I I’m doing internal linking on a page, I want to link to all of my very similar, semantically similar pages. Also my pages that are giving users the next steps to something. And you’re thinking, “Oh, that’s obvious.”
But think about it. You want to link your best resources on this and related topics. You also want to link to the user of next thing maybe further down the buyer journey or next questions they likely will have. Okay? That’s what you want to do within your site. Or if you own multiple sites when you crosslink. You don’t want to link externally to a page that has the exact same intent or very similar intent, because what you’re basically saying is, here’s my guide to email marketing.
The first thing you should do is go to this other website that has a guide to email marketing. Super bad. Don’t do that. You want to link externally to appropriate adjacent sources. So flip that on its head. What you want to do is find pages that represent authority and thought leadership on adjacent topics where you would be a logical, non-competitive source. When you do that, that’s how you build outreach programs.
So start looking at those topic models a little different. If you’re in market niche, flip ’em upside down on the words that you want. If you go to the bottom, look at the pages ranking for those things and the variance. That’s how you turn topic modeling AI into outreach strategy. We don’t have a software for that yet, but I see the advance. It’s a super advanced use. It’s a super advanced use case. I’ve pitched it internally many times. It’s such an advanced use case that internally, I have a hard enough time getting it done. Maybe I’ll do a blog post about it, “How to use MarketMuse for link building”.
Rich: So let’s talk just for a bit about using AI for improving a piece of content. This is something that we all need to do occasionally. We’ve got a great piece, but then, after a while it just becomes old, it becomes outdated, there’s more competition. It starts to get less search traffic. What are some of the ways that a tool like MarketMuse can help us improve and reverse the trend of a download trending content piece on our site?
Jeff: Yeah, great question. So fundamentally basics of optimization from a content quality perspective. What you’re trying to look for is, if one we’re covering a topic very well, specific topic, and they were truly the expert, what are the concepts that they would naturally cover with this particular piece. And potentially maybe a small collection of pieces. That should be part of your thought process when you’re optimizing. That it shouldn’t probably be only an existing single page. It should be that page and their brothers and sisters in the cluster should probably all be touched at the same time.
But if you are updating one page, that’s critical. So what are you looking for? You’re looking for maybe blind spots that you had that were topics that are logical to an expert. So an expert would clearly know how to integrate those concepts into the word. You’re looking for historical over optimization. Maybe somebody saw a list of keywords and copied the fact that Amazon was in the search results and then ended up copying a lot of components of the Amazon page or doing something that they should not have done. So you’re looking for some signals of over optimization.
There are software products that inform you on how to over-optimize, which is great for our clients. I love it when their competitors are over optimizing because it’s so obvious. You can tell why, and you can destroy them every time if you’re thinking thoughtfully. So optimizing, including those topics elegantly.
And then I like to say there’s a story in a list of keyword variants, or the topic model. And your article should be able to look at this list of words and turn it into that logical story. And omit, don’t include things that don’t fit with that story. And the skill that one can develop is being able to look at a topic model or a keyword variant list, understand where you have authority and not already, and know which ones to include and ways to elegantly include them to make your piece as good as it can be. And then double check against the competition and make sure you’ve got differentiation and it’s higher quality.
Those steps, it might manifest… The way that I described it is very abstract, but it might say, hey, you were writing an article about content strategy, and you didn’t include anything about buyer personas or target audiences. Please include sections on buyer personas and target audiences in your article and here’s a couple other items that would illustrate stuff about personas and audiences that would round out that exhibition of expertise. Here’s some ways you can use those things. And then when you’re done with that update, you can check it against who’s performing well in organic search, say my article includes those concepts – which an expert would agree – and it’s more comprehensive than anyone I’m competing with. I’m putting my business’s best foot forward.
So the first part is differentiation, comprehensiveness, and relative comprehensiveness. If you can do those three things, you optimized your page right. And if you do it for your cluster and find ways to make sure they’re woven together logically, that’s level two.
Rich: Excellent. This has been really helpful. And if you’re listening to this and you’re feeling a little bit lost, some of these topics can feel a little bit ephemeral. My best recommendation is sometimes just play around with the software. Once you start using these tools, some of these ideas start to fall into place.
And with that kind of caveat I’m going to ask Jeff, if people want to learn more about you, if they want to learn more about MarketMuse and how to use it, where can we send them?
Jeff: So jeff@marketmuse.com, you can shoot me an email. I’m active on Twitter and LinkedIn. My Twitter is @jeffrey_coyle and LinkedIn is @JeffCoyle.
And certainly play around make it a real workflow. Don’t just type in kitty cats, right? Look at the last page you create. Look at the intent, Get specific and analyze that text, and the first time you see a blind spot you had, that’s when it clicks. It really does. And that’s when you’re saying, oh, wow. Yeah, I did forget that. Because we’re fallible and your expertise doesn’t match others. So when the first time that buyer persona in the content strategy article, or you realize, oh yeah, it would make my page better if I included mentions of that. That’s a big difference than what you’re going to hear in the industry, right? Where it’s like, throw in all these words. That’s not what I’m telling you. And so just keep that in mind when you go through it.
And yeah, go check out MarketMuse. Our standard offering, you can buy it self-service, you don’t have to talk to anybody. But if you are a team and you’re making decisions about what content to create and update, if it would make a meaningful impact if that batting average went up, please reach out to me and I’ll make it work.
Rich: Awesome. Jeff, this has been great. I really appreciate your time and your expertise. Thanks for coming by today.
Jeff: Ah, thanks so much, Rich. I appreciate it.
Show Notes:
Jeff Coyle helps his clients make creating quality content a worry of the past, by utilizing AI solutions and SEO as part of the content strategy. Head on over to the MarketMuse website to see the solutions they‘re offering, and be sure to follow Jeff on Twitter and LinkedIn.
As President of flyte new media and founder of the Agents of Change, Rich Brooks brings over 25 years of expertise to the table. A web design and digital marketing agency based in Portland, Maine, flyte helps small businesses grow online. His passion for helping these small businesses led him to write The Lead Machine: The Small Business Guide to Digital Marketing, a comprehensive guide on digital marketing strategies.