Mini App Builder — Episode 4: 57% Shared. Zero Paid. The Honest Numbers After Two Weeks.
Two weeks after launching SpeedTap on Telegram, here's where things stand. Total users: 79. Monetag ad revenue: $0.001. Telegram Stars revenue: $0 (one test purchase I made myself to verify the flow works). Total spend: $10.
By any conventional metric, that's a failed project. Revenue to date wouldn't cover a coffee. So why am I writing this post?
Because one number in that data set is doing something interesting. And if you're thinking about building on Telegram Mini Apps, that number matters more than the others.
What This Post Covers
The full data from two weeks of running SpeedTap — Phase 1 MVP plus Phase 2 multiplayer features combined. User acquisition costs through AdsGram, ad revenue through Monetag, the user behavior metric that's genuinely promising, and the ones that aren't. This is the honest accounting that the first three episodes have been building toward.
The User Acquisition Experiment
Organic discovery on Telegram was going slowly. I ran a small paid test to see if user acquisition worked as a lever. Budget: $10 on AdsGram, the Telegram-native ad platform.
The campaign got rejected the first time with no explanation. After some digging, I changed the target language from "All" to "English only" and resubmitted. Approved on the second try.
Results:
Impressions: 3,237
Clicks: 168
Click-through rate: 5.19%
Cost per click: $0.06
New users acquired: 77
$10 for 77 users means a cost per install of about $0.13. For context: typical mobile game CPI ranges from $0.50 to $5+ depending on vertical and geography. AdsGram delivered users at roughly 2-10x lower cost than traditional app store ads.
Add 2 organic users who found the bot through Telegram search, and that's the 79 total users I had to work with.
The Number That Matters Most
Here's the one I keep coming back to.
Share click rate: 57%.
Out of every 100 users who completed a game, 57 of them tapped the share button. For comparison: typical mobile games see share rates of 5-10%. Apps designed specifically for virality often hit 20-30%. Hitting 57% means the share mechanic isn't just working — it's working more than twice as well as games built with massive marketing teams and user research budgets.
Remember Episode 2, where I designed the share message before writing the game? That decision was the biggest single bet I made on this project. The data suggests it was the right bet.
Monetag's CPM during Phase 1 was $0.34. This is below industry benchmarks ($2-6 for rewarded interstitials), but that's expected — the algorithm needs thousands of impressions before it optimizes ad selection. With fewer than 100 users, Monetag couldn't figure out what ads to serve. CPM would climb with volume.
What Didn't Happen
Now the parts that aren't working yet.
Telegram Stars revenue: $0. The admin dashboard technically shows 1 purchase, but that was me testing the payment flow to confirm everything was wired up correctly. Real user purchases: zero. With 79 users and Hard/Chaos modes also unlockable by inviting 5 friends, nobody chose to pay. That's a signal. At this scale, the "invite 5 friends" path is more attractive than the "pay 50 Stars" path. Or more likely: most users haven't even tried Hard mode yet because they're still working on Normal.
Daily challenge participation: lower than expected. Playing a game is easy. Coming back every day to play the same daily challenge is a different commitment. The daily challenge feature is built, leaderboards are live, but getting people to return for it consistently hasn't happened yet. This might be a retention problem specific to the current user base (most acquired through paid ads, less invested), or it might be the format itself needs more reward loops.
Retention data: too early to call on Phase 2. Phase 1 showed early signs but the user base turned over after Phase 2's feature drop. I'll have cleaner retention numbers in another month.
Cost vs. Return
The complete accounting:
Ten bucks down. Functionally zero in revenue. By startup math, this project has failed.
But startup math isn't the right frame. This series was never about getting rich off a tap game. It was about answering a specific question: can a non-developer build and launch a full-stack production app on a real platform, with real users, using AI as the coding partner — and can the numbers look anything but dismal?
The answer is: the code works. The users came (at $0.13 each). The share mechanism functions above industry norms. The monetization infrastructure is in place. What's missing is scale. And scale takes time.
If this project had cost $100/month to run, I'd have shut it down by now. At $0/month for infrastructure and $10 for the one ad experiment, I can let it breathe.
The Numbers
Full scoreboard after two weeks:
Total users: 79
Peak DAU: 14
Share click rate: 57%
AdsGram CPC: $0.06
AdsGram CTR: 5.19%
Cost per install: ~$0.13
Monetag CPM: $0.34
Monetag ad revenue: $0.001
Telegram Stars revenue: $0 (1 test purchase made by me)
Total spend: $10
Server cost: $0/month
Lines of code written by me by hand: 0
What's Next
SpeedTap keeps running. Apps Center resubmission is still pending review. Phase 3 is already sketched out — group challenges, weekly tournaments, ghost mode (race against your own best run), real-time random matching once DAU crosses 1,000. The 57% share rate suggests the viral loop is there; what's missing is enough fuel to light it up.
This episode wraps the current chapter of the Mini App Builder series. The pipeline is built. Two apps launched across two platforms for $10. The frameworks for future apps are in place — Toss on one side (pending business registration), Telegram on the other (live, monetizing, iterating).
The next episode in this series will come when there's a new data point worth sharing — Apps Center approval, the first real Stars purchase, a new app on a new platform, or some other signal that changes the calculus. Until then, SpeedTap is in the background doing what it does: waiting for people to tap 1 through 50.
← Previous: Episode 3: A Rejection Letter Became My Product Roadmap
More updates on the way. If you're working on something similar or found a smarter way to do it, drop it in the comments — the more we share, the faster we all move.
Disclaimer: This blog documents my personal learning journey. Nothing here is financial advice.
Comments
Post a Comment