Marketing for SMEs Podcast: Mike Montague on Real Stories vs AI Content

I joined Mike Montague, founder of Avenue9 and creator of Human-First AI Marketing, on the Marketing for SMEs podcast. Mike has spent 25 years across web design, radio, sales training, and podcasting. His take on AI cuts through a lot of the noise: use it to scale real stories, not manufacture fake ones.

• Watch on YouTube
• Listen on your preferred podcast app

Background: Three Threads

Jeremy: Welcome back to another episode of the show. I’m with Mike Montague, who’s been in this game for 25 years. He hosts a podcast called Human First AI Marketing, and his business is Avenue9.com. Let’s start with your background, Mike.

Mike: There are kind of three threads in my career that are interesting.

One, I started designing websites when I was in high school. Twenty-five years ago, that was a big deal. You had to hand-code it with HTML. But I didn’t want to be a nerd. I wanted to be cool and be on the radio.

So I was a DJ, hosted karaoke shows, was a top 40 DJ. I was Romeo on Mix 93.3, Kansas City’s number one hit music station. Had a blast with that. But it didn’t pay very well.

My dad was a professional sales trainer, so I moved from radio into advertising, into sales, then into public speaking and corporate training. That’s where I’ve spent most of my career. Hosting a podcast called How to Succeed for nine years, over 4 million downloads and 700 episodes. This is all three of those threads swirled together.

Jeremy: So your fascination with AI and podcasting, it’s not your first rodeo.

The LinkedIn Parallel

Mike: Right. I wrote a book published by LinkedIn about how to use social media for salespeople. And it was really about not spamming people. Just because we can reach more people on LinkedIn doesn’t mean we should. How do we use it to start better conversations and build our business?

When AI came out, I thought: oh no. This is the same thing, but on steroids. You could spam a million people, not just your connections. You could post a thousand times a day. So I felt like I had to take a stand and say, how do we do this the right way, with a human-first approach? I love all these tools. I’ve been a tech nerd since high school, but let’s use them for good, not evil.

Jeremy: When I watched your first episode back in December, you talked about three problems with AI: missing context, lack of creativity, and ethical concerns. Now that you’ve interviewed hundreds more people, has your thinking shifted?

Mike: Not necessarily on those three things.

The ethical concerns haven’t been addressed at all. There’s still bias in these tools. When you use AI, it’s predicting the most popular or most predictable answer, which, in America, that’s a middle-aged white guy like me. It predicts down the middle. It misses edge cases.

The context problem is still there too. AI is trained on all the world’s information, it kind of knows everything, but it doesn’t know which parts of that apply to your business. And that training was done three to six months ago, so anything new or novel, it doesn’t have context for.

Jeremy: I just read about the Synthetic Performers Law, just passed in New York three days ago. If you’re using AI and pretending to be human, you have to disclose it now.

Mike: That’s a step in the right direction. I saw a commercial selling coffee beans with an AI-generated spokesperson. I knew before they put the fine print up. You get that uncanny valley feeling. Why would you do that? There’s an actor. Have an actor read the script.

Jeremy: Advertisers want to test 16, 32 variations. They want to find the hook that worked and multiply it, that’s why they do it.

Mike: And there are some good use cases for that. Can we run the same commercial in 27 different languages? The company I worked for before I started mine operated in 27 countries and 19 languages, and I was the spokesperson. I could only speak English. So we had to have actors or trainers in each local market re-record everything. That’s a genuinely good use case.

The UGC Problem

Jeremy: With AI speeding everything up, people are now selling 40 pieces of UGC content for $40. It looks pretty real. To me, that kills the entire vertical. Then people go, I don’t know if that’s AI, I don’t believe it, I’ll scroll past it.

Mike: I have a strong view on this.

I call it my concentrated orange juice theory. A real user-generated story is like concentrated orange juice, you can water it down and spread it across all your channels. Use it on social, in your ads. It’s a killer story. It scales across everything.

But when you create an AI-generated version, it’s a weak version of that. When you spread it out, it doesn’t work. It’s not punchy. It doesn’t feel real. You’re ruining trust and burning those channels.

LinkedIn has tried to put things in place to stop AI from burning the platform down. Because if everybody posts spam and fake stuff, nobody reads anymore. Everybody’s posting, but nobody’s reading anything. Now that whole channel is burnt for everyone. You can’t stand out if everybody has fake stuff, how are you any different from your competition?

Jeremy: So people keep moving to the next thing. Reddit made a comeback. PR is coming back too.

Mike: They’re coming back because there are real conversations there. Not fake, real people saying good stuff and bad stuff. And podcast interviews like this are great because they’re real conversations on other people’s platforms.

I could create a video of myself and post it on my own YouTube channel. But AI and humans will only weight that with so much credibility because I’m the one saying it. When I come up on your show, or the New York Times says Mike knows about AI, that’s a lot more credibility. There’s a lot more trust from that.

The problem wasn’t a lack of content. We already had too much. The problem most businesses have is a lack of good content that resonates with people, that changes action and gets their attention.

What Guests Mike Brings On

Jeremy: Why does the podcast exist? Do you put people in buckets?

Mike: I filter a little. I should have more people on that I disagree with, but I don’t find them as compelling. What I try to do is find people with a similar value, building human-first but doing it in a completely different way.

James, who you mentioned works in 3D, world-building for video games. Has nothing to do with what I do, but he’s 100% in agreement that humans are creative, they do cool things. We need to make content for humans, not robots.

I’ve had data scientists, researchers doing sentiment analysis on micro expressions from audience reactions to commercials. The whole spectrum.

What I try to avoid are the people with commission breath. They’re ready with their talking points. I’ve had them on. I can smell it from their pitch and their website, hey, we can use AI agents to scrape a million emails and send automatic custom messages to all your prospects. Great. What do you do tomorrow? 

If you send a million spam messages to everybody who could possibly buy your thing today, what do you do next month, next quarter, next year? You’re going to run out of humans in a few days.

Viral Content, Organic vs Paid

Jeremy: You were talking about viral content in the pre-interview. In ads, everything lives forever as long as you keep paying. In organic, you need things to get traction.

Mike: There are three buckets you can put content into. Long-term, which builds over time. Then there’s organic-only, which I push back on, if your business is 100% referrals, you’re not trying hard enough. Imagine how fast you could grow if you ran a few ads.

And then there’s viral, which feels like the holy grail. A huge advertisement you’re not paying for. But that’s not how it works.

I’ve had multiple things go viral. One client was a rapper who got over 10 million views in the first week. His channel wasn’t monetised. The record wasn’t even for sale yet. He ended up with around 50 million views. And he basically made no money on it. Your chances of that happening again are so low, it’s not a sustainable strategy. It’s like hitting the lottery.

What I think about with content is: those viral spikes are nice little bumps. If I play the lottery every week, there’s a better chance I hit one. And 100 of the right people seeing a video is better than 100 million of the wrong people.

Your advertisements rarely go viral. It’s going to be a mistake you made, or a bad customer experience, or something funny that had nothing to do with your product. Really hard to monetise.

So what I try to do is create a sustainable drumbeat, build an audience over time that adds up. If I get ten people a week, a year from now I have 500 listeners. The system builds on itself. It builds reputation.

If you show up week in and week out, it’s the reliability that builds trust. Builds authority. And they get to know you authentically over time. One viral spike, there’s no emotional attachment. But if they hear you in their ear week after week, they start knowing you and trusting you. If you’re also giving generously, interviewing experts, sharing thought leadership, giving to the audience, those four things build loyal clients over time.

What Avenue9 Actually Does

Jeremy: As a service, what does Avenue9 actually help businesses do?

Mike: I’ll interview the founder, the top salesperson, the best clients. I’ll collect those stories, then use AI systems to repackage and distribute them on different channels. I call it scalable storytelling. It builds trust over time on whatever channel makes sense for your business. We monitor the channels, monitor the conversations you’re having in your business, and create content to build out the marketing plan.

I love podcasting, but I’ve worked with an IT company where they don’t want to be on camera. So I’ll talk to their customer success person and ask: what dumb things did people do with their technology this week? Who fell for a phishing attempt? Who wasted money on a product they didn’t need? We collect those stories and turn them into marketing.

It’s basically three things: the strategy, the story we’re going to tell, and the AI systems to make sure it’s not cumbersome for the client.

Jeremy: And they need to be at a certain size?

Mike: Usually a million or more. If you’re a solopreneur under a million, you kind of have to figure this out yourself, you can listen to me and Jeremy and learn. Over $30 to $100 million, they probably have internal teams. 

My sweet spot is the one to five million range. Too small to hire an internal specialist, but something needs to happen. It’s more fun to help people grow and make an impact, help businesses stay employed or hire more people. And at that size, you’re working directly with the owner. No politics. Once it gets to PE firms and investment, it gets messy fast.

Key Takeaways

• AI-generated content is a weak substitute for the real thing. Real user stories are like concentrated orange juice. AI versions are diluted from the start, they don’t carry the same trust or punch, and spreading them across channels accelerates the damage.
• The viral content strategy is essentially a lottery. Sustainable audience growth through consistent, generous content creation beats waiting for a spike that can’t be monetised anyway. Ten new listeners a week compounds into real results over a year.
• Commission breath is easy to spot. Scraping a million emails and blasting them isn’t a marketing strategy — it’s a one-day solution that burns every channel you have. Sustainable marketing means thinking past tomorrow.
• Avenue9’s model is built around story extraction, not content manufacture. Rather than generating content from thin air, Mike interviews founders, salespeople, and customers to collect real stories, then uses AI to distribute them at scale. The authenticity is the asset, AI just handles the logistics.

---