YouTube Video

Watch on YouTube

Category: Content Marketing

Tags: AI Content MarketingApp PromotionHuman ConnectionIndie App DevelopmentSocial Media Engagement

Entities: 11 LabsAICling 2.1FA.AIGemini Flash 2.5InstagramLearn Piano appTikTokYouTube Shorts

Building WordCloud ...

Summary

Transcript

I set out to answer the simple question, can I promote my apps using automatic content marketing generated by AI? I spent over $1,000 building an app to test the latest models and the results, 14 likes. 14 likes. And that's when I realized AI content marketing is actually worse than we think. In this video, I'll share what I learned after three months of building, testing, and failing, and the three hard lessons you need to know before you waste your time and money on AI content marketing. I'll cover why my AI marketing attempts failed, the hidden cost of automation, and the secret weapon for growing your app using content marketing. Right now, indie developers are successfully promoting their apps on Tik Tok, generating millions of views and building entire app empires from short form content marketing. And why not? You can get access to millions of potential users, and it doesn't even have to cost you a single dollar. So, of course, I wanted to automate it. And with my experience growing a small YouTube channel, I've learned a lot about video pacing, hooks, and what gets traction. Maybe I could combine that knowledge with AI to create the ultimate content creation tool to promote my apps. First, I had to choose what content strategy would be easiest to automate with AI. And there are plenty of storytelling formats you can use to promote your app. There's reaction hooks, slideshows, point of view videos, and case studies to name a few, but I went with the reaction and hook format because it's the easiest to automate. You show a one to five second reaction shot of a person with an intriguing hook on screen. The objective here is to connect within instance with the viewer. Stop them from scrolling and you only have like fractions of a second to actually make that happen. Arguably, it's the most important part of your video. And the app itself pulls context from the app store and from the video that you upload. The idea is to create a video that is relevant to the target audience and promotes the core feature of the app itself. I used Gemini Flash 2.5 for video context and script writing, 11 Labs for audio, and I used FA.AI to bypass 11 Labs consecutive access issues. And I used the latest model of Clling, which at the time of recording this is Cling 2.1 for video generation. It's quite an expensive model, but it looks the most convincing. And in total, I spent $1,1226 generating this AI and testing it to build my app. You select an AI influencer, add your demo recording, and within a few moments, you get a video ready to be posted to Tik Tok. And the result, it's okay. But here's the real test. Will the generated videos even get any views? To test, I'm going to use a real account for my Learn Piano app. It has a small following on Instagram, Tik Tok, and Shorts. It's what we call a warmed up account because, as it turns out, social media platforms don't give you instant virality. In fact, they have a sophisticated system that will watch your app usage to determine whether you're a legitimate user or just a marketer looking to spam their platform. So, if you're a marketer looking to spam their platform, you need to act like a real user. For the first couple of weeks, you need to use your account like a normal person would before you even post a single video. Just scroll through and like things. This is what's called a warmed up account. And I've had thousands of videos from previously posted content. So, it should be really easy to get traction with AI generated content. Right? Wrong. I generated hundreds of videos using my apps and I selected the best ones, the ones with the most compelling hook and the most intriguing story lines, but they just didn't get any traction. On Tik Tok, the most views any video got was 440 and on Instagram 249. YouTube Shorts, on the other hand, looked really promising. In the week that I tested, I generated 13,000 views with the top video generating nearly 2,000 views, which looks really good. But when you dive a little bit deeper, you see the engagement is just really low. There's barely any likes or comments there. And this is my most popular video. 77.2% of viewers decided to simply swipe away. And it's a similar story across all of the AI generated content on this channel. Meanwhile, the same content, same channel, has a 90.3% like to dislike ratio from my previous content. And that content was just me showing my app while I'm sitting on the piano and playing. So, why is this content getting less views? 4 seconds into the video, retention is good, sitting at 102%. That's really good. And there are three noticeable audience drops. First, directly after the hook, when the demo starts. second when the app is referenced and third when the final call to action displays. But engagement is low. Only six likes with no comments. And Tik Tok is even worse. 20% retention after 4 seconds with a steady drop throughout the whole video. And this seems to be the common trend across all AI generated videos that I posted to Tik Tok. I've spent countless hours and sleepless nights working on this app. For the past few weeks, I've listened to nothing but AI generated audio. And I can wholeheartedly say that AI generated audio just sucks. I don't think your brain processes it like normal speech. For the past few weeks, I felt a bit off. I've had this growing sense of tension and uneasiness. And I can't say for sure, but I think that the excessive exposure to AI generated audio has contributed to this weird feeling that I've had. And it wasn't until I decided to stop work on this app that the feeling started to subside a little bit. And now when I hear AI generated audio, I feel slightly repulsed. And it's kind of happening on a subconscious level. And I can't explain why exactly. And maybe that's happening for viewers, too. Maybe they just don't like it. And they don't know why. They just don't like it. And if that's the case, no script changes or hooks. Nothing like that's going to fix it. As a comparison, let's have a look at my best performing video on the channel. It was uploaded earlier in the year and it has no generative AI. It's just me playing the piano with my app sitting in view. 7,100 views, 14 subscribers, 108% retention after 4 seconds with 64.3% of viewers choosing to watch instead of swiping away. Not amazing results by any standard, but a good comparison. I stopped posting content on this channel and I stopped promoting my piano app. Instead, I wanted to focus on building a tool to automate marketing, but my videos were kind of getting traction at the time, and people were starting to subscribe. We're not talking massive numbers here. It's just a nice honest start. And that's when it finally hit me. I wanted to build a shortcut, a quick and easy way to churn out content to promote my app. I thought, this is a problem I'm having, so other developers are also having the same problem. So, I focused on creating a solution, an app that automates the process entirely. But there's no shortcuts in life. If I had have kept at creating content for the channel the old-fashioned way with a human being, me sitting in front of the camera, I would have learned a lot more about what works and what doesn't. I would have continued to amass followers. I would have continued to grow views and I would have continued to connect with viewers who wanted to download my app. And it kind of does remind me a little bit like playing the piano. I never expected to have immediate results. It's a slow journey that only works when you put the time in, when you grow, and when you learn. When I use AI to automate any process in my life, I'm finding that it's just robbing myself from learning any new skill. In this case, I'm robbing myself of learning how to promote my app, how to grow a following, and how to get more comfortable promoting my apps in short form content. And that's kind of the key. The missing piece in any content marketing growth strategy is the human element. And as cliche as that sounds, as an indie app developer, I have the flexibility to work on any project I want to work on. I can follow my passions and I can build anything I want to build. And if I focus on building a project that I love and I'm passionate about, it's easy to create content to share that journey with others. And then it becomes less about selling and more an exercise in attracting others to me. Creating content and sharing my passion to others who have the same passions. In a world where Tik Tok is filled with AI generated content, our biggest strength is actually us. authentic human connection and experiences. Think about that just for one moment. If every single video promoting an app is AI generated, people are going to just scroll past it. But when they see a real person having a real experience, that is the advantage. So yeah, the app was originally built to automate my content marketing, but I ended up learning a whole lot more along the way. I don't think I'm going to continue working on this one. I'm going to shelf it. I'm going to let it sit and brew for a bit. I might revisit it down the path, but until then, I'm going to take my own advice and build the thing that I'm actually passionate about. This learning piano app.