A client came to me with a familiar problem:
“We built something people love. But growth is slow. What if we run ads?”
They weren’t new to building — just new to paid acquisition. And like most early-stage teams, they weren’t trying to burn a pile of cash on five platforms just to “see what sticks.”
They needed speed, clarity, and results. So we rolled up our sleeves and got to work.
Here’s what worked (and what didn’t) — so you can skip the guesswork.
Don’t start with Google Ads unless you love chaos
They kicked things off with Google because, hey, they had a $600 free ad credit. Seemed like a no-brainer.
Wrong.
Performance Max gave them a $0.05 CPC at first. Looked amazing. But the conversions? Crickets.
Even worse: Google only shows real conversion data after 24 hours. So you make changes, scale budgets — then wake up to a $7 CPC and no users.
We shut it down fast.
Lesson:
Free credits are bait. Google is great if you know what you’re doing, but if you’re early-stage with no clear conversion data, skip it.
Meta Ads pulled the most weight
This is where things got interesting.
We ran Facebook/Instagram ads — but only for desktop. (The app didn’t work on mobile, so no point wasting impressions.)
We also made sure every image looked good per surface. Meta throws your ad across multiple placements — feeds, stories, sidebars — and if your image is stretched or cropped, it tanks performance.
We used Canva to create proper versions for each one. Took an extra hour. Worth it.
Then we segmented campaigns by country. US CPC was $1.50. Brazil, Mexico, Philippines? As low as $0.02.
So we turned off the US and scaled what was working.
Result:
This became our highest-performing channel. It drove the majority of our 10K users — at some of the cheapest rates I’ve ever seen.
Extra tip: If you’re running Meta ads, watch your placements like a hawk. Start with automatic placements to collect data, but then split things out. Instagram Feed will behave differently than Facebook Stories. And if you’re not measuring per placement, you’re guessing.
TikTok didn’t convert — but here’s why
We boosted a TikTok video that the founder had lying around. $5/day.
It got more views. But that’s it.
Why? Because TikTok doesn’t reward lazy content. You have to actually create for the platform. Record, test, iterate. Then boost what performs.
We didn’t do that — so it flopped.
Lesson:
Don’t throw money at bad content hoping it’ll work harder.
Also: TikTok needs a volume play. Not one or two videos. Ten to fifteen variations minimum. It’s a creative testing engine — not a plug-and-play platform.
Reddit Ads had potential, but the platform didn’t show up
The Reddit Ads UI is amazing. Super clean, super flexible.
But the ads never left “review mode” for an entire week. Support didn’t respond. So we gave up.
You can import Meta ads into Reddit with one click, which is smart. But if the ads don’t run, what’s the point?
Lesson:
Maybe worth trying later for them. But Reddit isn’t a good bet if you need fast results and actual support.
Newsletter ads: Expensive? Yes. Effective? Also yes.
We tested a few AI/Design newsletters — ranging from 50K to 300K subscribers.
When the issue dropped, traffic spiked. After 24 hours? Dead flat.
But the quality of those users was better than anything we got from social. More engaged. More curious. Fewer tire kickers.
Also — newsletter operators will give you a discount if you ask nicely and sound organized.
We sent them a clean Google Doc with our copy and creative. If you don’t, they’ll just write something with ChatGPT — and it’ll suck.
Lesson:
Great for credibility. High-quality bursts. Just don’t expect a consistent drip of users afterward.
If you’re going to try this route, pick newsletters that match your target user. Don’t just chase list size — ask for open rates, click-through rates, and past advertiser screenshots if they have them.
Tracking: How we knew what was working
We didn’t fly blind. Every campaign had clean UTM parameters, and we used GA4 to track what people did after they clicked.
What we tracked:
- Sign-ups
- Activation rate (actually using the product)
- Country and device splits
Most importantly: we set benchmarks in advance. What does a good CAC look like for this product? What’s the LTV of a typical user?
You don’t need fancy dashboards. But you do need numbers.
What actually got us to 10K users
We didn’t get fancy. We just did these things well:
| Channel | Worked? | Why / Why not |
| Google Ads | ❌ | Conversion lag + bad CPC spike |
| Meta Ads | ✅ ✅ ✅ | Desktop targeting + cheap CPCs + good creatives |
| TikTok Ads | ❌ | Boosted bad content instead of testing properly |
| Reddit Ads | ❌ | Never ran. Support ghosted us |
| Newsletter Ads | ✅ | Expensive, but great quality + visibility |
Final advice if you’re about to run paid ads for a consumer app
- Start with Meta. Low CPM, great targeting, and forgiving enough to learn on.
- Don’t spend a dime until your conversion tracking is dialed in.
- Budget small, test fast, and kill losers quickly.
- Build creatives per platform — lazy resizing will cost you.
- If you’re not mobile-friendly, don’t run mobile traffic. Obvious but overlooked.
- Boosting content = fuel. Make sure the engine’s working first.
- Track everything, even if it’s in a spreadsheet.
- Don’t run ads just because you’re bored of slow growth. Run them because you’re ready to scale what’s working.
If you’re sitting on a great product and wondering if paid ads are worth trying — they are. But only if you’re willing to treat it like a channel, not a slot machine.
Want help figuring it out? Get in touch.