Developer | Building apps to $30k/mo
New York, NY
Joined February 2025
- Tweets6.2K
- Following305
- Followers658
- Likes8.6K
On of my users this morning ran into my app's accountability feature twice, then cancelled their trial.
For my app over the past week I'm finding ~50% cancel their trial for no apparent reason, and ~50% cancel right after running into friction somewhere in the app.
Yesterday, after checking posthog with chat, I found a large portion of my users are cancelling right after they experience a bug with one of my AI features. I fixed the bug, added additional graceful degradations in case it happens again, and submitted an expedited app review.
Checking posthog again this morning, I found this user that just cancelled their trial right after running into visible friction (pressing a button multiple times) with an accountability feature.
And they say you should talk to your users, which I should, however thanks to modern analytics + AI it's easy to find instances where users are cancelling due to friction in your app and what the friction is that should be resolved.
I got the most trials today (7) I've ever gotten in a single day
(Day 55 of building apps to $30k/mo)
But just cause I'm getting trials doesn't mean they're converting...
TL;DR:
- About one week into a $50/day meta trial starts campaign
- My first 2 meta conversions resulted in billing error
- Unfortunately a really high trial cancel rate from meta (~80%) - much higher than with apple ads and organic (~30%)
After filming this though I realized a large portion of trial cancels are happening right after a bug users are experiencing in one of my primary features. I submitted an expedited app submission and still waiting for review. Hopefully this fix can provide a good conversion lift
My users coming from Meta (~20% trial -> paid) have been converting far worse than users coming from Apple Search Ads and organic (~70% trial -> paid)
At first I thought Meta was just giving me low quality users, but after digging into the posthog data it looks like there's a much more explainable (and fixable) reason...
It turns out Meta users use one of my app's AI features about 3x more than users from Apple Search Ads or organic. This makes sense because my Meta creatives are built around this particular AI feature.
It *also* turns out that there's a bug that shows up ~25% of the time with this AI feature...
About half of my Meta users who cancelled their trials used this AI feature, encountered the bug (I can see via posthog), some of them repeatedly, then immediately cancelled their trial.
I fixed the bug and submitted an expedited App Store review (still waiting).
Also goes to the importance of verbosely littering analytics events throughout your app to understand why users are cancelling or even why they're converting.
I got the most trials today (7) I've ever gotten in a single day
(Day 55 of building apps to $30k/mo)
But just cause I'm getting trials doesn't mean they're converting...
TL;DR:
- About one week into a $50/day meta trial starts campaign
- My first 2 meta conversions resulted in billing error
- Unfortunately a really high trial cancel rate from meta (~80%) - much higher than with apple ads and organic (~30%)
After filming this though I realized a large portion of trial cancels are happening right after a bug users are experiencing in one of my primary features. I submitted an expedited app submission and still waiting for review. Hopefully this fix can provide a good conversion lift
I got the most trials today (7) I've ever gotten in a single day
(Day 55 of building apps to $30k/mo)
But just cause I'm getting trials doesn't mean they're converting...
TL;DR:
- About one week into a $50/day meta trial starts campaign
- My first 2 meta conversions resulted in billing error
- Unfortunately a really high trial cancel rate from meta (~80%) - much higher than with apple ads and organic (~30%)
After filming this though I realized a large portion of trial cancels are happening right after a bug users are experiencing in one of my primary features. I submitted an expedited app submission and still waiting for review. Hopefully this fix can provide a good conversion lift
Today is a record day of $170 active trials for my iOS app!!
(Day 51 of building apps to $30k/mo)
Current revenue: $383/mo
TL;DR:
- $170 of active trials atm - the most I've ever had in one day
- looking like the meta campaign MIGHT be profitable!! (been 3 days so far of this campaign so need to let it mature; running trial start campaign at $50/day)
- all with AI-generated videos btw - I think AI UGC is 10000% the future. Far cheaper, far more control, and even higher quality than most human UGC you will find
PSA: my apple search ads CRs (tap-to-install rate) increased ~50% across campaigns after adding an app store header image
Today is a record day of $170 active trials for my iOS app!!
(Day 51 of building apps to $30k/mo)
Current revenue: $383/mo
TL;DR:
- $170 of active trials atm - the most I've ever had in one day
- looking like the meta campaign MIGHT be profitable!! (been 3 days so far of this campaign so need to let it mature; running trial start campaign at $50/day)
- all with AI-generated videos btw - I think AI UGC is 10000% the future. Far cheaper, far more control, and even higher quality than most human UGC you will find
When meta finally starts understanding who to target
Today is a record day of $170 active trials for my iOS app!!
(Day 51 of building apps to $30k/mo)
Current revenue: $383/mo
TL;DR:
- $170 of active trials atm - the most I've ever had in one day
- looking like the meta campaign MIGHT be profitable!! (been 3 days so far of this campaign so need to let it mature; running trial start campaign at $50/day)
- all with AI-generated videos btw - I think AI UGC is 10000% the future. Far cheaper, far more control, and even higher quality than most human UGC you will find
Drew retweeted
Polsia is about to hit $25M run rate.
Next stop: $100M ARR.
And things keep accelerating:
$0 → $1M in 12 weeks
$1M → $10M in 10 weeks
$10M → $20M in 6 weeks
I just published one of my favorite and most successful aisloP episodes yet: “The Run Rate.”
I tell the whole story, from idea to $20M, month by month.
The episode ends in Tokyo, where I went a few weeks ago.
Probably the best place on Earth to take a deep breath, get inspired, and rethink what’s possible.
I came back with an even bolder vision for Polsia.
The road from $25M to $100M is going to look very different.
Polsia v2 is coming soon.
smh i guess these serial trial starters are a thing huh
Today is a record day of $170 active trials for my iOS app!!
(Day 51 of building apps to $30k/mo)
Current revenue: $383/mo
TL;DR:
- $170 of active trials atm - the most I've ever had in one day
- looking like the meta campaign MIGHT be profitable!! (been 3 days so far of this campaign so need to let it mature; running trial start campaign at $50/day)
- all with AI-generated videos btw - I think AI UGC is 10000% the future. Far cheaper, far more control, and even higher quality than most human UGC you will find
I never knew how drastically conversion rates can vary by acquisition channel.
For example with my iOS app:
- Apple ads - this is where I started and conversion rates were pretty good. 10% converting to free trial at paywall
- Organic - barely any would even install. I got organic TikToks and Instagram videos to thousands of views, clearly shouting out the app, and would get like 2 installs which could have even just been organic ASO??
- Meta ads optimized for app installs - the cheapest app install I've gotten. Got it as low as ~$0.50 per install on one of my creatives. But these users converted like shit. Actually a 0% conversion rate after 2 days of this meta campaign
- Meta ads optimized for trial starts - the highest CPI I've had yet - even up to 2x CPI of Apple ads, but starting trials at 2-4x the rate of Apple ads.
It's crazy how an organic Instagram Reel can have horrendous install rates, like 5k impressions -> 1 install; and the same Reel in an ad campaign optimized for installs can have 300 impressions -> 5 installs.
My CPI has been varying a lot tbh - the first day I was getting CPI of ~$1.50 which should have been barely profitable given my previous metrics of install -> trial start rate, but none of these users were converting (cause it was optimized for app installs)
The last 3 days on the trial start campaign I've been getting much higher CPIs - like $5-7 - but these users are converting at almost 20% on one of my paywalls - like 2x the trial start rate of any of my other acquisition methods
The campaign is currently getting better every day (only day 3 rn of this one), and it's already looking like it may be profitable as of yesterday, but hopefully it will optimize even further cause it's just a little over 1 ROAS as of now. So unless metrics were to suddenly drastically tank I'm going to leave it up for a few more days and reassess then
Today is a record day of $170 active trials for my iOS app!!
(Day 51 of building apps to $30k/mo)
Current revenue: $383/mo
TL;DR:
- $170 of active trials atm - the most I've ever had in one day
- looking like the meta campaign MIGHT be profitable!! (been 3 days so far of this campaign so need to let it mature; running trial start campaign at $50/day)
- all with AI-generated videos btw - I think AI UGC is 10000% the future. Far cheaper, far more control, and even higher quality than most human UGC you will find
Day 49 of building apps to $30k/mo
It's day 2 of my first trial start meta campaign - 3 successful trial starts (all three canceled tho) and the trial starts cost too much. CPMs through the roof and CTR on the floor.
I ran an app install campaign a few days ago and CTR was good - 2-4% after the first day. These metrics have not transferred to this new trial start campaign.
Holding out hope that this is just the optimization process so planning to leave these running at least for the next few days...
Plz comment if this was similar to your experience / looks like this will optimize further or not
I know this is not that much money.
But I have NEVER had $130 in active trials before.
Gotta celebrate the small wins 🙏
Day 49 of building apps to $30k/mo
It's day 2 of my first trial start meta campaign - 3 successful trial starts (all three canceled tho) and the trial starts cost too much. CPMs through the roof and CTR on the floor.
I ran an app install campaign a few days ago and CTR was good - 2-4% after the first day. These metrics have not transferred to this new trial start campaign.
Holding out hope that this is just the optimization process so planning to leave these running at least for the next few days...
Plz comment if this was similar to your experience / looks like this will optimize further or not
This is the difference between optimizing for App Installs vs Trial Starts on a Meta campaign:
October 12 I was optimizing for App Installs and did get far more installs, and even 16 IAP transaction starts, but no one started a trial.
(the 1 user who did was one of the few installs not from the meta campaign)
October 13 I created a new campaign optimizing for Trial Starts. I'm getting half as many installs, even half as many transaction starts, but far more Trial Starts.
(Most of my traffic hits in the late evening-night, starting right about now, so hopefully Oct 15 metrics will bump up too)
Day 49 of building apps to $30k/mo
It's day 2 of my first trial start meta campaign - 3 successful trial starts (all three canceled tho) and the trial starts cost too much. CPMs through the roof and CTR on the floor.
I ran an app install campaign a few days ago and CTR was good - 2-4% after the first day. These metrics have not transferred to this new trial start campaign.
Holding out hope that this is just the optimization process so planning to leave these running at least for the next few days...
Plz comment if this was similar to your experience / looks like this will optimize further or not
RIP fellow paid ads gang.
Did I choose the worst time to start experimenting?
Day 49 of building apps to $30k/mo
It's day 2 of my first trial start meta campaign - 3 successful trial starts (all three canceled tho) and the trial starts cost too much. CPMs through the roof and CTR on the floor.
I ran an app install campaign a few days ago and CTR was good - 2-4% after the first day. These metrics have not transferred to this new trial start campaign.
Holding out hope that this is just the optimization process so planning to leave these running at least for the next few days...
Plz comment if this was similar to your experience / looks like this will optimize further or not
A lot of people on this app saying to not provide free trials.
Reminder: THERE IS NO HARD AND FAST RULE
I dislike free trials because it makes testing times longer - you have to wait for the trial to end to see if the user actually converts or not
But, for this most recent app, I discovered free trials convert FAR FAR better. Makes sense - all my top competitors have free trials
Day 49 of building apps to $30k/mo
It's day 2 of my first trial start meta campaign - 3 successful trial starts (all three canceled tho) and the trial starts cost too much. CPMs through the roof and CTR on the floor.
I ran an app install campaign a few days ago and CTR was good - 2-4% after the first day. These metrics have not transferred to this new trial start campaign.
Holding out hope that this is just the optimization process so planning to leave these running at least for the next few days...
Plz comment if this was similar to your experience / looks like this will optimize further or not
Why is it so anxiety inducing starting a new meta campaign?
I don't know if it's so much about potentially losing money as it is about potentially being faced with the truth that your app is not profitable yet... not ready to scale yet... and not exactly knowing what it is you need to change to make it ready
It's day 2 of this new trial start campaign and I *think* it's looking healthily on it's way to optimizing further over the next couple days. Will see. Or it's back to the drawing board... 🫡
Day 49 of building apps to $30k/mo
It's day 2 of my first trial start meta campaign - 3 successful trial starts (all three canceled tho) and the trial starts cost too much. CPMs through the roof and CTR on the floor.
I ran an app install campaign a few days ago and CTR was good - 2-4% after the first day. These metrics have not transferred to this new trial start campaign.
Holding out hope that this is just the optimization process so planning to leave these running at least for the next few days...
Plz comment if this was similar to your experience / looks like this will optimize further or not
Testing pricing as the trials come in from meta - $29.99/yr vs $39.99/yr
Curious to see how these compare over time
Day 49 of building apps to $30k/mo
It's day 2 of my first trial start meta campaign - 3 successful trial starts (all three canceled tho) and the trial starts cost too much. CPMs through the roof and CTR on the floor.
I ran an app install campaign a few days ago and CTR was good - 2-4% after the first day. These metrics have not transferred to this new trial start campaign.
Holding out hope that this is just the optimization process so planning to leave these running at least for the next few days...
Plz comment if this was similar to your experience / looks like this will optimize further or not
Day 49 of building apps to $30k/mo
It's day 2 of my first trial start meta campaign - 3 successful trial starts (all three canceled tho) and the trial starts cost too much. CPMs through the roof and CTR on the floor.
I ran an app install campaign a few days ago and CTR was good - 2-4% after the first day. These metrics have not transferred to this new trial start campaign.
Holding out hope that this is just the optimization process so planning to leave these running at least for the next few days...
Plz comment if this was similar to your experience / looks like this will optimize further or not
Day 48 of building apps to $30k/mo
Current revenue still $371 (not for long)
Yesterday I got the MOST installs in one day I've ever had (35), all thanks to meta ads. BUT - I was optimizing for trial starts and NONE started a free trial (more below)...
TL;DR:
- Most installs in a single day yesterday (35) but none converted
- Meta campaign wasn't letting me optimize trial start (figured it out - now they are - started new campaign will see how it goes 🤞)
Day 48 of building apps to $30k/mo
Current revenue still $371 (not for long)
Yesterday I got the MOST installs in one day I've ever had (35), all thanks to meta ads. BUT - I was optimizing for trial starts and NONE started a free trial (more below)...
TL;DR:
- Most installs in a single day yesterday (35) but none converted
- Meta campaign wasn't letting me optimize trial start (figured it out - now they are - started new campaign will see how it goes 🤞)
