| Source | https://www.youtube.com/watch?v=0YHKGcHetOA |
|---|---|
| Readwise URL | https://read.readwise.io/read/01kw1f9mdsp7dahjychjrcmwbc |
| Readwise ID | 01kw1f9mdsp7dahjychjrcmwbc |
| Author | Adam Lyttle |
| Category | video |
| Site | YouTube |
| Published | 2026-04-22 |
| Saved | 2026-06-26T08:03:46.745000+00:00 |
| Tags | adam-lyttle, app-development |
Readwise Summary: I paid $2000 to figure out Meta ads so you don’t have to. Here’s every mistake I made going from $36.87 per install down to $6.87… and the one creative format that actually worked for my indie app.
Follow my journey here: Website: https://adamlyttleapps.com Twitter: https://x.com/adamlyttleapps Github: https://github.com/adamlyttleapps Instagram: https://instagram.com/adamlyttleapps TikTok: https://tiktok.com/@adamlyttleapps Substack: https://adamlyttleapps.substack.com
Shout out to my amazing video production team at https://clipwing.pro/
#indiedev #indiehacker #buildinpublic #appdev #metaads #facebookads #appmarketing #aso #vibecoding #solodev #startup
$36.87. That’s what it cost just for one person to tap install, to tap a single button. And my stomach dropped when I saw that number. 3 days and $2,000 later, I now have that down to $6.87. This video is everything I got wrong in between, so you don’t have to. And by the end, you’ll know the mistake that burned through $500 in a single day, the simple creative strategy that reduced my
install cost by five times, and the one thing nobody tells you about paying for ads that you only learn when you’re already bleeding money. This is the playbook I wish someone had handed to me a week ago. If you’ve built apps and only ever relied on App Store optimization, stick around. Quick context, I’ve shipped a lot of apps, and I make a living as an indie app developer. And until last week, I had never paid a single dollar for marketing on Meta ads. Every user I’ve ever had
came through free traffic sources, mostly App Store optimization and keyword research, showing up when someone was actually searching for what I had built. I have no growth team and no actual budget. It’s just me, and I got comfortable. A weak screenshot, a clunky paywall, onboarding that drags, who cares? The traffic is free. Some of them convert, and the ones that don’t didn’t cost me a single dollar anyway. But free traffic has a ceiling. And last
week, I was flying to Tokyo to give a talk at Try! Swift. 12 hours in a seat with nothing to do felt as good of a time as any to finally figure out this whole Meta ads thing for my new app, Spech. And every single assumption that I had had was wrong. Mistake number one. I thought Meta ads would be like App Store ads. With App Store ads, you add your app, you set your budget, you pay, then Apple uses your App Store
screenshots as the creative, and shows your app within the App Store. But with Meta ads, it’s different. You have to bring your own creative, the actual video or image that someone sees when they’re scrolling Instagram or Facebook. And nobody actually tells you where to start. So, I did the obvious thing. I recorded my screen using the app, stitched together the best moments, and just used that. It’s pretty much what App Store ads does, so why not? I mean, how bad could it be? $36.87 per install. Ouch. At that rate, I would
need every person that clicked my ad to subscribe for 2 years just to break even. This doesn’t feel like marketing. This feels like I’m actually throwing money out the window. Clearly, screen recording just doesn’t work. So, I did what every marketing guru on the internet told me to do, and that led me to mistake number two, UGC, user-generated content. You’ve seen these before, real people filming themselves using the app like they just
discovered it. Every marketing swears by it. So, I went to Use Clip, paid $400, and I was matched with three creators. Two delivered great videos, and the third one kind of just ghosted me. That’s a whole different video. Now, I’ve got a professional video I can use for my Meta ads creative. And the install cost dropped from $36.87 down to $17 for this video and $11 for this video, almost three times cheaper.
This is the same app. It’s pretty much the same messaging. It’s just not a screen recording. Okay, now we’re getting somewhere. The game is about the creative. Make it better, pay less per install. Got it. Now, I have a new problem. Every new hook, every new angle costs me another $200 to pay for user-generated content to be created. If I wanted to test five different angles, five different hooks, five different audiences, that’s $1,000 and weeks of
waiting before I even know what actually works. And that’s assuming the creator doesn’t just run away. So, I tried something simple, a slideshow, or as Meta calls it, a carousel. There’s no actor, and there’s no filming required, just a few photos with a single line of text. I asked Claude to come up with the text, “Your best memories are buried on your camera roll.” And I slapped it onto a screenshot of my actual camera roll. This is so basic, so simple, I was actually a little embarrassed to test
it. $6.87 per install. The best result so far. Aha, there’s something here. I don’t need to record an entire video just to test a hook. I can spin up 10 different ideas in a single afternoon, and it won’t cost me anything. Then run them all, see which one lands, kill the losers, and double down on the winners. And if I want to, I could pay to get a bunch of user-generated content created using that hook, knowing it has a much
better chance at converting into actual installs. Turns out, I was doing Meta ads all wrong and spending a lot of money in the process. Instead of testing with user-generated content to see what works, I’m now testing with slideshows first. But to get any meaningful results from Meta ads, you need to set up Facebook SDK in your app. This is new to me. Here I am again, naive, thinking this is just advertising. Like, I just pay to show an app, not integrate
tracking code into my app from Facebook. Wrong again. At this point, I’m traveling around Tokyo, and any free moment I have in the back of a cab or on the bullet train, I’m juggling kids, I’m preparing my Try! Swift talk, and I’m trying to get Facebook SDK integrated. And there was this one point where I just nearly gave up. This is completely out of my league. How do others make this look so effortless? Am I actually on the right path here? Will this even
work? Am I just throwing my money away? There was a moment, a brief moment, where I actually did stop the campaign. I gave up. But something told me to keep going, just one more experiment, one more try. And that’s when I made the slideshow, out of sheer desperation, with 3% battery left on my MacBook in the back of a taxi. If I quit, I would have walked away thinking Meta ads just doesn’t work for indie apps. And I think that’s exactly where others have quit, trying and failing, not knowing what the
answer is. So, I’m making this video today to show you how I finally flipped the script and saw that there actually is potential here. But reducing the cost of app installs is just part of the story. Only one in 10 people who install the app are actually signing up for a free trial. Even at $7 per install, every trial sign up is costing me $70. And of those trials, only a fraction will actually convert into paying users. It’s too early to tell how many will,
but I guarantee it won’t be 100%. Back when I was paying nothing for users to download my app, I didn’t care. But all of a sudden, it’s important. It’s the difference between my app failing or thriving. So, how do you actually make this work? From what I can see, there are actually three levers. Lever number one, get the cost per install down. You do that by optimizing the creatives and getting the right target market. Lever two, increase those trial sign-ups. Do that by fixing the onboarding, fixing
the paywall, and experimenting with your pricing. Or lever number three, and this is one I never really considered until I started paying for users. You drop the free trial altogether. All of a sudden, this option is making sense to me. I want to recoup those marketing costs quickly, and I don’t want to wait a week to see if the user converts. I want them signing up for an annual or lifetime plan right away. For Spech, my strategy is simple. Continue at the top of the funnel, work on getting users installing
the app, work on bringing down the cost per click. I’m aiming to get it around about $2 or lower. Then experiment with the onboarding flow. Maybe add some analytics to work out which steps have the biggest drop-off rate. And I can’t believe I’m actually saying this, but experimenting with a lifetime offer that has no free trial. Instead of a start your free trial button, I’ll just have an unlock now button. As always, I’ll keep you updated with the progress, showing you what works and what doesn’t. Subscribe, so you don’t miss out. I’m
learning that App Store optimization and Meta ads are not the same game. They require very different approaches and completely different pricing strategies. With App Store optimization, every person who downloaded your app was already looking for your app. They typed in the keywords, they had the intent, they were halfway sold before they even started using it. You just happened to be in the right place at the right time. But with ads, no one’s actually looking for you. Nobody cares that your app even
exists. Someone was watching a dog do a funny trick on Instagram a moment ago, now they’re looking at your ad. And it has to be good enough to not only spark their interest, but also convince them to open up their wallet and pay for your app. It’s a tough challenge. So, here’s my playbook for reducing the ad cost to $2. Build slideshows, test a bunch of hooks, find the one that lands, then pay a creator to amplify it. That’s it. Everything else is just optimization. If
you’re an indie app developer who’s only ever done App Store optimization, try a slideshow ad this weekend. Set the budget to $100, create a few different hooks, and see what happens. Today, my install cost is $6.87, but I’m aiming to get that below $2. Subscribe to see how.