Subscribe
Share
Share
Embed
Mike Montague, founder of Avenue Nine and host of the Human First AI Marketing podcast, explains why so much AI-generated marketing content turns into slop, and shares his framework for using AI to make marketing better instead of just faster.
Agency Forward explores the future of agencies as tech and AI drive down the cost of tactical deliverables. Topics include building competent teams, developing strategic offers, systemizing your business, and more.
New episodes delivered every Tuesday.
Chris DuBois 0:00
Hey everyone! Today I'm joined by Mike Montague. Mike is the founder of Avenue Nine and the host of the Human First AI Marketing Podcast and YouTube channel. He got his start in radio and sales training before running marketing for a large international company with over 200 offices across 29 countries, and now works with agency owners and business leaders on how to actually use AI in marketing. I got to meet Mike at a conference that we spoke at earlier this year, and I love the way that he talked about AI. He has tested a real practical framework for where AI belongs in the content marketing process and where it doesn't. So this, for me, is a relief because it's not just theory about the future of AI marketing. In this episode, we discuss why most AI-generated marketing content turns into slop and the strategy work that has to happen first to avoid it. The four buckets of deciding what AI should and shouldn't touch in your business. How gifts and invitations marketing works even better with AI-enabled personalization and more, no one was asking for another community, but I made one anyway. So what's different? The dynamic agency community is designed around access rather than content. Access to peers who've done it before. Access to experts who've designed solutions. Access to resources that have been battle tested, and right now the price for founding members is only $97 a year. Join today so your agency has immediate access to everything you need to grow. You can join at dynamicagency dot community. And now Mike Montague. It's easier than ever to start an agency, but it's only getting harder to stand out and keep it alive. Join me as we explore the strategies agencies are using today to secure a better tomorrow. This is Agency Forward.
Mike Montague 1:57
Why is most AI content slop, and how do you fix it? Well, I think there's a couple of reasons for bad marketing in general, and AI and slop just sort of accelerated it, right? So if we think back to why most marketing stinks, it's because there's no strategy to it. Somebody had a funny idea and thought it might make good content, or people are trying to pitch way too early, and they didn't identify their their target market, or they just are not graded all the different things that it requires to make marketing. So when you see a cheesy like local TV commercial, you're like, how did that ever get approved? Like, who wrote this script? Who are they targeting? It looks like they shot it, you know, like on an iPhone with bad lighting in their their auto dealership. Like that's just a bad commercial. And if you look at that, I think that's what's happening with AI now, and why there's so much slop is that everybody can make anything, and so it's like, oh, I can make it. Ask AI to make a six second video about a a you know car blowing up, and this is going to be the thing. And and so they have a bad idea. They do bad execution, and then they do sort of bad strategy or what I call like tying it all together. Like a lot of times, people even make good commercials, but there was no call to action, or there was it wasn't repeated enough, and then it turns into slop. Even though it was an interesting or good idea and aimed at the right person, it just doesn't come together into a marketing plan that's repeatable over and over and over again.
Chris DuBois 3:38
Right? Yeah, I think that that gate being now open for everybody to just create whatever they want definitely has. It was a floodgate, and so yeah, a lot of garbage comes through when there's a flood. And I saw this
Mike Montague 3:54
right away when AI hit. I was excited about it because I was doing marketing for a large international company. It has like 200 offices in 29 countries and 17 different languages, and I was like, "Oh, we could make a lot of really cool stuff. And then I just thought, I was like, "There's still some problems, though. This isn't a magic wand. We can't like fire all of our vendors and employees and just say, 'Hey, AI, like run our marketing for us. We still need to decide, you know, who we stand for, what our message is, what our unique brand is, where where's our voice, who are we targeting, and then we have to kind of figure out what the bandwidth of this is, and like what systems are we going to use, and and I could, you know, I can make an image now with something like 100 different image models for a campaign. Which 1am I going to choose, and which one's the right choice for me? And so there's this whole systems thing that really is a challenge, and getting all those systems to link together is a whole another challenge. And then also that comes. To this like bandwidth problem of the audience, what do they want to hear, and what do they want to consume? Just because I can make video easy doesn't mean any of my clients are watching videos,
Chris DuBois 5:11
right?
Mike Montague 5:11
So I can write blogs super fast, but people might not have the time to read it. Or if I make 1000 blogs in a day, I can make more than I can read myself, let alone my audience and a decision maker has time to consume. So there becomes some really important questions.
Chris DuBois 5:30
I think,
Mike Montague 5:30
especially for agencies, but but marketing leaders too. Of well, we need to have some discretion. We need to have some tastes, and we need to have a plan for this. Otherwise, it's all going to be slop,
Chris DuBois 5:41
right? Yeah, the blog one is is interesting because I saw it immediately with like working with agencies, right? There's so many SEO agencies that were just like, we can crank out content now, yeah. But it was it's not good content. It's also like if a like you're using AI to make the content, why doesn't someone just go to AI for the answer. Like they can skip the middleman here and just do their their chat function or whatever. But like I used it this weekend to help with with some blog posts. And but I realized so like I had it look through all of my client meetings, all of my prospecting calls, and give me titles. Hey, what are the problems people are talking about? What titles? I chose ones from that that I thought were the most pressing. Then I had to create questionnaires for each article, and then I just put that like open a Google Doc on my phone, answered each of those questions while walking around using my Whisper Flow. I got like 8000 sips doing this. It was probably pretty good, but it was like it was thorough for like documentation, and then I said, "All right, now let's turn this into some content. And so, still way faster than what you could normally do, right? But it's like,
Mike Montague 6:49
yeah.
Chris DuBois 6:49
But we're not just pumping out the slop and like hoping it sticks.
Mike Montague 6:53
That's what I got excited about too. So I sort of created a framework and a system for myself in order to make good stuff, and I call it the concentrated orange juice approach of like, if you take a bad prompt or a weak idea and ask AI to expand it into a full marketing campaign, it's going to water it down. You don't have like a strong enough concentrator, enough material there to to make a year's worth of marketing from, but if you've been in business for a while and you have a year's worth of conversations and you boil that down to a concentrated idea and say find the you know most interesting, most powerful moments from this year's worth of conversations with my clients and turn that into a blog post, that's really good stuff. Now you have that like really meaty, really interesting stuff that you can expand on, and it'll it'll be really great content.
Chris DuBois 7:49
And it only builds on that, right? Like if you if you've created your system, right? It's like when you then create the content, you get it just right the way like the way you want it. It now helps with the next like batch of content that you're going to create, because it knows what the good stuff is, and it's feeding those ideas. Yeah, you
Mike Montague 8:06
mentioned something else in your process that I wanted to highlight, which is you go back and forth. Way too many people try to do like that one shot prompt where it's like write a blog about this, and what I found works best is going back and forth and fighting with it and arguing and like brainstorming as a partner and saying, "Give me 10 ideas for the title. Now give me 10 ideas for the subtitle and an outline. And I like this one, but not that one. Or take these parts and then ask me five questions that you need to know before you write this blog article. And then write the article. And then I edit it and take out all the am dashes and the inner worlds, and it's not this, it's that, and and put my voice in it, and maybe a metaphor, and add some color or humor to it, and then I paste it back in. That's the part that you were just getting to. Is you got to create that, close that loop, and send it back, and say, here's all the changes I made. Here's my final draft. Learn from this for next time, and then over time it gets much much better,
Chris DuBois 9:05
right? And it's probably more more reason to to have your own like I don't not necessarily locally stored, but like a quad code instance that is like yours rather than just using the chat interface. That like everybody complains about the hallucinations of like ChatGPT and everything, but like if you have a specific context folder like on your computer that it knows like this is everything like just keep working with this like find those hallucinations disappear and you can just keep building it up and creating better stuff by having conversations. I think so
Mike Montague 9:40
too. I've also found that, like, as the models come out, like, they got them a lot more literal, and so sometimes you have to still update those regularly because it'll start taking that command where I said, like, never use an em dash, and all of a sudden it was removing the the hyphens between hyphenated words, and I was like, God, come. Like now, I gotta go back and fight with it again. And similarly, I've seen drift in some of those context files and things. That as I add to it, there starts to be like contradictions and stuff. Where with one client, I I wanted really lengthy, you know, narrative style answers, and the other one wanted really short bullet points, and now it doesn't know which one to do or which format to do when, and so I do still have to manage it kind of every time. I don't think there's any sort of like hands off like this is the the self improving system that's going to be set forever. It's really a yeah recursive learning
Chris DuBois 10:38
place. The weird process is still hard to hard to get. So let's talk about like what how you actually do help clients like through AI, and so I think there's a lot agencies can learn from this.
Mike Montague 10:51
Yeah, I think there's a couple of things that are interesting for me. One is in marketing, I switched to retainer based pricing right away, and so I realized quickly, like you said, that if I'm charging for blog posts, I can't charge per post anymore. They would be they'd owe me $50,000 by the end of the day, and also I can't charge per hour because I could create 1000 blog posts in an hour, and then I would be the one getting hosed. I wouldn't be able to make enough money or stay busy enough. So I learned pretty quickly that I needed to switch to some sort of value-based pricing and just say, you know what, I'm going to solve AI marketing problems for you. This is what I charge per month to do that. If that's great, I don't have contracts or anything, but I just go monthly, month to month after a three month minimum, and I just say, as long as you're still getting value and I'm still solving problems, we'll keep going. You can stop anytime, and that's really allowed me to move into these three buckets, which I think is what you're actually asking about, which is sort of the the The human first AI strategy bucket, which is again, even if people can do it themselves, they don't know what to do necessarily, or they don't have creative marketing ideas. And in any business, it's hard to see the forest from the trees. My dad always used the term "you can't read the label from inside the bottle. Like it's just really hard to do marketing for yourself, even if the AI tools make it easy. So a lot of people hire me for the outside perspective. Number two is the AI marketing systems. They're just AI is such a huge thing. They can't. It's got to be a team sport. They can't keep up with all of it anymore. So do they know that you know Claude released an update that made it better at this than ChatGPT, or that Gamma is a really cool tool that they've never heard of before, or that now there's a new integration from Claude into to PowerPoint, and so you might not need Gamma for your presentation anymore. You can go straight there. So a lot of them enjoy kind of talking about the tools and the systems and the AI part of getting marketing automation to work together, and then the last part is the kind of traditional agency content stuff, which is I interview their founder, their top salesperson, their best clients once a week or once a month, and create the marketing content for the plan of of doing the process that you just said, and say, hey, what's give me five ideas for a blog post or internal company videos or whatever we happen to be making? I'll do that creation process for them because they don't really know how, and then they're not great at it. That's not what they love doing. I've have all kinds of different businesses, but one of them is an IT company, and so he knows all about IT. He knows nothing about creating marketing, right? So he doesn't know what stories are interesting, what questions to ask, what a good blog post would look like when it came out, and so I just interview him and take care of all that for him, but he knows all the IT systems integration part. I have another client that is a marketing agency that works with with healthcare businesses, and they know all about marketing. They could do it all, but they don't know the technology side. So I do the consulting and systems set up for them to be able to do what they want to do,
Chris DuBois 14:22
yeah, and it's interesting because like the guy who's really good at IT might take AI and say, "Well, now I'm equipped to be able to create all of these marketing assets and just produce more of that slot. Right? Like even if it does have like it's super smart on the IT side, it doesn't mean the marketing side is going to actually do what they need,
Mike Montague 14:42
I'll say. The hard part is convincing people of that because I also have clients that think they're like, "Oh, well, Claude said our marketing should say this, or our website should say this, and I'm like, Claude didn't say anything. Claude is not a person. Claude doesn't have an opinion on whether your website should be in dark. Or light mode, it doesn't know. We're gonna have to like you know test that, or we're gonna have to use human intuition here and stuff. And so resetting that is a large portion of that first bucket of strategy that I talk about too. Is it's really easy for people to get fooled because the AI will tell them it does, right? So it'll tell that IT guy, "Oh, we know how to do marketing, and they'll just say, "Act as a marketing professional and write an email for me.
Chris DuBois 15:27
Now you said that that
Mike Montague 15:28
doesn't cut it. That's not not good enough. We can talk about why if you want, but I'm sure you've covered it before. Yeah,
Chris DuBois 15:33
no, I would love to get into it. I want to share it quick. When I was redesigning my website, the guy that I was talking to, he said, "Yeah, have to talk to Claude about some of these design changes. And oh, I didn't realize Claude could do all the design changes. And he's like, "Oh, hang on, not AI. My team member's name is Claude. It's a person. I'm like, "Oh, that's really unfortunate for this guy, like forever. But yeah, I have
Mike Montague 16:00
a cousin named Alexa, and it's annoying. Even her mom's like, "Alexa, bring me a cup of coffee. It's like, "Don't do that. Yeah, I think the the thing people miss here, and it's an important part of the discussion, is that AI is trained on all of the world's information, and and mostly available on the internet, and we all know that that's not very good. But I came from a world of sales, and of all the salespeople in America, over 50% of them miss quota.
Chris DuBois 16:34
Right,
Mike Montague 16:34
they they're not very good. And so, if you ask AI to write a sales message or a 30-second commercial for you, based on what the average salesperson would do, you are going to miss quota. It's going to be bad, and there are some things that you can do to kind of get it up to like 80% if you ask really specific or you give it context and and stuff that it can do actually better than the average person. But simply. simply asking it to pretend to be an expert in that doesn't work. And then I think to get it to 100% you still need the human involved. You need somebody that's done the role and understands the the context, the situation, knows when to break the rules, knows when somebody would get bored, and and there's just there's a different level of expertise that AI has not reached, and and I don't know if it ever will.
Chris DuBois 17:30
Yeah, it's a great point about the sales themes, and even looking like larger scale, I just read something about like like half of the content that we're seeing right now on the internet is created by AI, and so it's like so. If we're using we're using AI to create more content without like the the human in the loop here, like helping with giving it this stuff, it's like we're just creating the same thing over and over. And so now, like, what are we actually doing, right? Like,
Mike Montague 18:00
yeah, it becomes a photocopy of a photocopy, and it just kind of degrades over time. And now I'm old because I don't even know that people photocopy things. But I think that like it makes sense, right? AI can't train on AI data because it starts like this loop of of drifting where it just kind of gets slowly gets worse and worse over time unless you reset it and give it some real world context. I know you use a ton of real conversations like this one with the podcast and say, "Hey, these were two real humans talking. Learn from this, learn their language, learn these ideas. You know, study this and then make something. And we can kind of keep pulling it back on track a little bit, and we can still make good stuff. But it's interesting. I think a lot of people are trying to go too far too fast, and there's a lot of hype and discussion about what it will be someday. And it's really hard to sort out what it can do for you and your business today. And today, it's a really nice intern that doesn't know anything about anything, but it's got a genius level IQ, and so you can say, "Hey, intern, here's what I want you to do: go do that. And then you got to check the intern's work and and you know make sure it's doing it and hold it accountable, and coach it, and train it up, and give it context, and all that kind of stuff. And eventually, someday we might get a a nice employee that can, you know, know everything we know about the business. But today, that's not where it is. And so many people, I think, want to jump ahead to. I just put my business hit the easy button in the morning, and it gets all my leads and closes them for me, and and then does all my work, and I just take the check. That that's not where we are.
Chris DuBois 19:48
Yeah. So I actually want to get into how how you're thinking about context, because like one of the so like bringing up the podcast when I'm working on any of these blog posts, like my. My cloud code instance has access to all of my podcasts, the transcripts of everything, and so sometimes I'll be talking about one article and stuff, like giving it my answers. And then when I look at the written version, it actually pulls out things and says, you know, when I was interviewing Mike on on this podcast, you know, he said this, and it's like able to connect all of those dots that even I might not necessarily like. I don't remember every podcast I've recorded, but it's pulling these conversations. How are you thinking about like context? I don't know, like documentation, consolidation, any of these things in order to get better results.
Mike Montague 20:36
I think that's exactly the kind of thing that humans can't do. So we met at an event where we were both speaking on stage. So you've heard me do this bit before. But there's kind of four buckets of things in my mind. There's things that like humans should always do. If it's like the magic of your business, if it's relating to another human being, if it requires empathy that or or the importance that you show up. The joke I always make is like, if it's you know the anniversary dinner with your partner spouse, you might want to show up for that. Don't send your AI avatar. Like maybe it could hold a better conversation than you will at this dinner, but the point is not to do that. The point is that you show up, and in your business, there's certain things in that bucket. There's also things that like you humans shouldn't do. I use the examples of like hazardous mining or you know bomb disposal or something. If we can use an AI or a robot to go do that, great. We should definitely do that. There's things that like humans can't do in this AI bucket too, which is you're saying, I I used to pay people overseas to like listen to every episode or transcribe it. I started podcasting like I don't know, 12 years ago now, 15 years ago, where like I had to have somebody listen to every word and type it out and stuff. So it's possible, but it's way slower. No human should be doing that work, and somebody had to tag all the keywords in the podcasts and and stuff like that. But the two buckets, I'll do the other bad one, and then we'll talk about the good one. There's a bucket of things neither person should be doing. There's a lot of just busy work, and AI shouldn't do it or can't do it yet, or humans can't do it, and I think you as a leader need to think about what should nobody be doing. Just because we can do all of this extra busy work or reporting or data doesn't mean we should. I see people waste weeks of their life coming up with a report, you know, in Cloud Code or something that they they didn't even really need or isn't going to inform any decisions in in their business, especially in smaller business. A lot of people start tracking or doing like eye tracking on their website and stuff like that. You're not there yet, you know. You don't you don't have millions of visitors, and you're trying to maximize e-commerce sales and and stuff. Like, don't waste your time on that. That's something you shouldn't be doing, but the biggest bucket is what I call like the human plus AI bucket, and this is my Iron Man analogy of like Tony Stark has Jarvis AI in his ear; he's got the suit around him, it turns him into a superhero, and a lot of things that fall into that bucket are the stuff that you're doing and the stuff that I love doing, which is how do we look at mass amounts of data that we couldn't otherwise process and pull out insights from? How do we get creative or speed up like things that would make humans super bored at doing things like reminder sequences for unpaid invoices, or or if you do events and stuff, you know that's just annoying to make six different emails. It's like our event starts in one hour, our event starts in one day, our event starts next week. Like doing all of that is just super boring. There's also some really cool like additive stuff we can do, and I talked about this a little bit before too. But creating for different personas, or foreign languages, or even personality traits-like there's some really cool stuff that AI can do that we can't do.
Mike Montague 24:16
I can't speak Spanish, so my available my podcast isn't available in Spanish unless I have AI do that, and those are the kind of things where I think we need to focus more on, and not this like pipe dream of someday I hit the easy button, but like what could I do that's super cool that nobody else is doing right now with these tools? You've come up with a lot of those. What what are some of your favorites, or or what have you always wished AI would do that it it either can or can't do? Yeah.
Chris DuBois 24:47
So one of the ones that um okay, so I built a tool for um for my clients specifically, but it's to track all of our marketing. Like tests that we're running, and so before it was done in spreadsheets, and it was just like it was hard to manage, and like, hey, client, like this is this is with you now, and like going back and forth. Now it's built as a tool. With I used AI to I don't know how to code, so like I used it to do that. But it can take all of the results from all tests across all of my clients, it'll anonymize this and stuff, but it'll look at for pattern matching across the tests that we're running. So, like, hey, for people with this type of ICP, we're actually seeing that these things are changing, and now I've got this like proprietary data to to realize what's happening on the front lines of a marketing, and I can take like a new client and say, "Hey, we need to be doing these things right now because I I've seen that they're working for your ICB, yeah. And and that one, like just that, is like I could never have done that before. It would have taken so much time sitting down and looking across all my clients, like, "All right, what's working? What's not? Which I would have done, but it just the time, you know, to get there. Yeah, for me, it's
Mike Montague 26:01
a lot of like personalization too. That like I would have loved, and sometimes on big accounts, I would take the time to like go into my pitch deck and say, "Oh, let me change the slide, or let me put their logo on this, or or let me you know come up with some slightly different messaging because you know I want to make sure this is perfect. Those are the kind of things I think we have the speed now. It's not that I'm revolutionizing my business. I'm just making what I was already doing that much better, and and that's kind of what I hear from you. Is like if you had the time and availability, you would have gone through their you know transcripts and found the data or found a cool way to help your clients, but you just weren't able to do that with the bandwidth and tools available two years ago, and so a lot of that stuff is changing. And it's fun to to see which ones actually do work now.
Chris DuBois 26:50
One of my buddies has he's working large tech company, but he built a database that will take every sales conversation and every customer support call gets updated, but this the system will find the persona that they're talking about. It'll assign it to that conversation, and then it'll look for any like any other information about like what features are they talking about, what questions are they asking, and so it'll surface objections and everything. But like their CEO was able to look at the data and said, "Oh, we need this feature right now. Like this is coming up in so many of the conversations. So they like moved, you know, the roadmap changed and they prioritized that and brought more sales. Yeah, but they're also seeing like what do people care about within this based off the questions they're asking? What do they care about in this space that we might not see except through these questions, and so now they they can actually use that information to stay ahead of everybody else in the market. Yeah, it's crazy how you can you can do these.
Mike Montague 27:51
Agree. I think those are really cool, and especially for agency owners. I think we've known this for a while, but I've always said that there's like there's two kind of secret hacks for me in in marketing, and they're gifts and invitations. That if you try to pitch a message or you try to like go out there and sell something, people are going to be adverse to that. They're going to push back. Kind of hard selling and hard marketing is just good for like hard pushback and and resentment. But if you offer somebody a free gift, they generally don't say, "Oh, I can't believe they gave me something. Like how rude of this company to give me a a gift. And also, if you invite somebody to a party or a group or or some sort of experience that is, you know, something they wouldn't otherwise be invited to. That's different because even if you don't go, you're happy to be invited, right? So it's a little like hack into the the psychology of the buyer. So if we think now with AI, like what makes a great gift, and I always think about like Halloween candy. Is like, are you going to be the house that's like the please take one? It's like, oh, we can make a bunch of AI slops. So I'm just going to throw a bunch of candy out on the porch, take whatever you want. Like, sure. Sometimes it's fun to hit up that that house once and like you know scoop all their candy out. But nobody respects those people. Nobody nobody really likes that. There's also like the full size candy bar house, right? And so, okay, well now with AI, I can make something really good. So now I could be like the full size Snickers house, and I could be giving away like something that people go, "Hey, we got to remember that address right down the blue house on the corner next year. We're coming back for that because that was really good, but there's also this level of gift giving now where it's like we could know what Chris's favorite candy was, and he's not a Snickers person; he's a Reese's Pieces person. So now I'm going to do this custom Reese's Pieces with Chris's name on it, so it's Chris's Pieces. And it's going to have this awesome packaging, and now we're talking a whole other level. If you're like, "Oh my, I didn't can't even believe this existed. I owe you one for this. This is amazing. That's next level marketing that really gets people's attention, and they're not going to turn off. They're going to sign up. They're going to tell other people about it, and it's so remarkable that they didn't even know it was possible. That's the kind of stuff that I like to focus on. I think that's the stuff that's available with AI now.
Chris DuBois 30:28
Before moving to my next question, I need to thank you for saying Reese's pieces because convincing my family that it's not Reese's pieces is one of the biggest frustrations that I have. I thought you were a
Mike Montague 30:43
cups guy. I thought you were you're anti cups, but yeah, that's great.
Chris DuBois 30:47
So, I guess what what's like something that you've built that you're just like super proud of. You're like, man, the the the ability to like build this is was mind blowing for me, and now I have this as a resource that I can just deploy. Is there something that just jumps out at you that you're super proud of?
Mike Montague 31:06
Not necessarily on like the the deploy front. I think you're much better at that than I am. I think about some content. So my heart's always been in like the content and the messaging and the the storytelling part of it, I've been an interviewer. I mentioned podcaster for a long time, and one of my clients is a an old folks home senior living community. Sorry, they'd be mad at me already in Kansas, and they they have a bunch of employees. They they didn't really have the money or bandwidth to create like employee training videos. But I was able to interview the owner and their like director of nursing and and people in their community to to create some training videos for them, and then use AI to read the AI voices. Like they didn't want to be on camera, and they didn't have like the budget and and stuff to do that. And these videos turned out really good. The voices in in Google now are just amazing in Google Vids, and we did the slides in Gamma using AI tools, and we used the transcripts from Zoom in the meeting to do that. And we kind of used like four or five different AI tools, but the outcome was on budget, on time. We cranked them out, and then the videos, like I was getting a little teary. Like we're talking about like saving people's lives if their medications messed up, or a safety issue, or like training somebody how to treat treat a person with dignity. That's you know moving to a senior community and the challenges with that-like they really were doing really good stuff-and it's like, man, that's something that they couldn't have afforded
Chris DuBois 32:51
to do two
Mike Montague 32:52
years ago. And now we're solving a problem that's like making a difference in people's lives and saving lives in you know a few hours with a few 1000 bucks. That's that's pretty cool.
Chris DuBois 33:02
Yeah. So, what's the 8020 here? If an agency needs to focus on like those few things that are actually going to move the needle for them, where would you point them?
Mike Montague 33:17
I think it goes back to deciding what to make first. So my tagline for my podcast is the human first AI marketing, and and same thing for my company. And I think you got to do the work first. You got to decide what you stand for, who you're building it for, and and what you're trying to accomplish before you go make anything, and then I think like from there, it's also kind of human in the middle and human last. But but that human first message to me is like you do the thinking first. You put your customers first. A lot of people I see being too greedy, and they're saying, "Oh, this could save me 10 hours a week. But what if it costs your clients 50% more time, or headache, or stress? Like that's not a good decision. So customers first too, and I think that eliminates a lot of slop and spam and other cheesy things that people would do. And finally, human first to me means employees first. Just because you can automate somebody's job doesn't mean you should. And I think you can do more together than you can by replacing it with the system, or free that person up to do better work. And too much of the conversation is aimed at job loss or replacing work and things that I don't think is helpful. So, I put all of that into my human first bucket as my 80% If we if we focus on those things, I think we're going to be doing the right things.
Chris DuBois 34:50
Awesome. Cheers. I can get behind that. Last two questions for you. With the the first being, what book do you recommend every agency owner should read? The
Mike Montague 34:59
book. Gift the most is by Seth Godin, and he is the self-proclaimed, I think, world's greatest marketer. And I recommend you check out all of his stuff. But the book "What to Do When It's Your Turn" and then it's in parentheses, and it's always your turn, is a really cool book. It was a compilation of some of his blog posts, but it's a magazine format, and it's all aimed at taking responsibility and like what to do when AI or disruption comes into a market, or what to do when you start your own business, or what to do when there's one of those inflection points and you need to step up. And there's some really cool motivational and intelligent things that'll make you think, make you feel, and help you move forward in your life and career.
Chris DuBois 35:47
Yeah, based off your description, it gives me like war of art kind of vibes, which I know Godin was a fan of as well. But yeah, yeah, check it out. One of his few books that I haven't even heard of so. Last question: Where can people find you?
Mike Montague 36:06
Well, it's not very hard. I'm an internet marketer, so if you type Mike Montague or Avenue Nine into whatever search engine AI thing you're doing, you'll find me. But Avenue nine.com and the Human First AI Marketing Podcast is the best places LinkedIn and YouTube are my channel. So Mike Montague on LinkedIn, or YouTube is Avenue Nine AI, and there's like 75,000 subscribers over there. So you can join, check out the podcast, and check out some good clips for marketers trying to use AI.
Chris DuBois 36:36
Awesome. Well, Mike, good seeing you again. Thanks for joining.
Mike Montague 36:39
Thanks.
Chris DuBois 36:44
That's the show, everyone. You can leave a rating and review, or you can do something that benefits you. Click the link in the show notes to subscribe to Agency Forward on Substack. You'll get weekly content, resources, and links from around the internet to help you drive your agency forward.
Unknown Speaker 37:00
Forward.
Transcribed by https://otter.ai