I helped this nutrition tracking iOS app go from losing
$15,000/mo
to making
$60,000/mo in profit.
Immediately.
Meta ad spend vs cost per purchase
Jan 25 – Apr 12 · daily · audit Mar 17
Lukas Szanto
01 / 10 →
From losing $15K/mo
to making $60K/mo in profit.
Immediately.
Daily profit & loss (after ad spend).
Daily profit / loss (on ad spend)
Jan 25 – Apr 12, 2026 · audit performed Mar 17
Lukas Szanto
02 / 10 →
Slide 03 · The setup
What they had when I started.
A nutrition tracking iOS app with real product–market fit and a $2,000,000 ARR.
And an agency running their Meta ads into the ground.
→Every campaign optimized for Highest Volume bidding.
→Broad audiences with no creative segmentation.
→No real structure for testing creatives.
The agency's answer to unprofitability?
"Let's try a different audience."
Lukas Szanto
03 / 10 →
Slide 04 · The diagnosis
You can't media-buy your way out of unprofitability.
Once you have the fundamentals of campaign structure and media buying down, there's almost nothing you can do inside Meta Ads Manager that moves the needle.
—Not "optimizing" audiences.
—Not stacking bid rules.
—Not "letting the algorithm learn."
If the campaigns don't work, it's one of two things:
1
The creatives aren't good enough.
2
The post-click funnel isn't good enough.
That's it. Everything else is theater.
Lukas Szanto
04 / 10 →
Move #01 · Bidding
1What I did first
I switched every production campaign to cost-per-result bidding.
The spend dropped immediately.
That scares most people - and that's exactly why most people won't do this.
But here's the truth:
The drop in spend IS the point.
Cost-per-result bidding only spends when Meta can get you a purchase at the price you set.
And that's what you want.
When the campaign isn't spending it means one of two things: Either your creatives are weak, or your funnel is.
Lukas Szanto
05 / 10 →
Move #02 · Testing
2How I test creatives
I rebuilt creative testing with bid caps, not Highest Volume.
On Highest Volume, a single terrible ad can eat thousands of dollars before you realize it's a loser.
With a bid cap set at ~2x your target cost-per-result:
✕Losers get no spend — they can't compete at the bid.
✓Potential winners get a real chance.
✓You can push 10, 20, 50, 100 new creatives per week through without chaos.
✓Testing went from an expensive guessing game to a cheap, scalable process.
Lukas Szanto
06 / 10 →
Move #03 · Creative
3What came next
I started producing creatives at volume.
Once testing is cheap and structured, the bottleneck moves.
It's no longer "how do we test ads efficiently?"
It's "how many good ideas can we bring to the table each week?"
→New concepts, new angles, new formats - tested daily.
→Winners got scaled.
→Losers didn't get spend.
No more loser ads burning cash for three weeks before someone notices.
Lukas Szanto
07 / 10 →
Slide 08 · The result
That's what flipped the account.
Monthly P&L on ad spend
–$15K→+$60K
Cost per purchase
~$130→~$50
No new ad budget. No "secret hack."
Just stopped lighting money on fire in testing, and forced the ads to actually earn their spend.
Lukas Szanto
08 / 10 →
Slide 09 · What's next
And we're not done.
Every ad in this account still points to the homepage.
That's the single biggest lever left on the table.
Next 90 days
→Dedicated landing pages per ad angle.
→Testing homepage vs. direct to app store vs. quiz funnel vs. advertorial.
→A/B tests running at every step of the post-click journey.
If the ad account flip was a 5×,
this is the 10×.
Lukas Szanto
09 / 10 →
Slide 10 · Your move
Want me to look at your account?
I audit Meta Ads accounts for iOS subscription apps.
Same approach. Same process. Same kind of findings.
Just applied to your specific situation instead of someone else's.
How to reach me
If you want an audit, connect with me and send me a message.
DM me on LinkedIn.
Lukas Szanto · Profitable subscription app growth
10 / 10 ✓
Tweaks
Scroll through slides with ← / → or tap the edges on mobile.