A story about how buying traffic for Telegram Mini Apps stopped being an experiment and became a steady income stream. No “secret schemes”, just media buying, analytics, and working the sources.
Adexium Team | March 2026
Let’s get straight to it: arbitrage on Telegram Mini Apps isn’t a “money button.” Our team spent months fine-tuning setups before we hit stable volumes. The result: over $500K in revenue and an average ROI of ~185% on our top project.
How’d we pull it off? We stopped spraying and praying. Below is how we bought traffic at Adexium, which creatives actually converted, and where we burned budget in the early days.
Where We Started
We ran traffic to two of our own TMA products (acting as direct advertisers):
- Case A: Hypercasual clicker (tap-to-earn with potential token drop)
- Case B: PDF utility inside Telegram (subscription-based)
Simple goal: acquire active users as cheaply as possible, the kind who’d hit /start, invite friends, and buy premium inside our mini-apps.
“We weren’t hunting for magic bundles. We just stared at dashboards every day and ruthlessly cut zones that were pumping bots instead of real users.” Alexandra, Head of Media Buying.
Case A. Clicker, 185% ROI over 4 months
Tap-to-earn projects have a short lifecycle. We needed to build an audience before the token listing. Geo focus: emerging markets, 90% Android.
Results (Nov 2025 – Feb 2026):
- Spend: $176,430
- Revenue (monetization + In-App): $502,847
- Profit: $326,417
- Average ROI: 185%
Top Geos: Brazil, Indonesia, Philippines, Nigeria, Vietnam.
How We Scaled:
| Month | Spend | Revenue | ROI |
|---|---|---|---|
| Feb 2026 | $61,430 | $219,847 | 257% |
| Jan 2026 | $60,000 | $171,000 | 185% |
| Dec 2025 | $40,000 | $94,000 | 135% |
| Nov 2025 | $15,000 | $18,000 | 20% |
Month one was all about testing and building blacklists. Once we cut the fraud, ROI started climbing.
Case B. PDF utility in Telegram, ~99% ROI
A long-game product. Audience: students and office workers. User LTV is higher, so we bought traffic more carefully, via In-Page Push.
Results (5 months):
- Spend: $94,230
- Revenue (premium subscriptions): $187,450
- Average ROI: ~99%
Top Geos: USA, UK, Germany, Canada.
Key insight: iOS traffic converted to premium purchases 45% better than Android, even though clicks cost more.
Creatives that got clicks
Users are tired of cookie-cutter CTAs. Here’s what actually worked for us:
In-page push (Case B, utilities)
- System alert style: Gear icon or warning badge. Copy: “Storage full. Compress heavy PDF files?”
- Personalization: “Your {device_model} supports new Telegram features. Update now?”
Popunder (Case A, tap-to-earn)
- Gamified pre-lander: Simple wheel or treasure chest; click to “unlock” 10,000 coins. CTA: “Claim bonus in Telegram”
- Social proof: Fake chat interface where users “brag” about cashing out
S2S tracking for Telegram
Without solid tracking, TMA arbitrage is just gambling. The main headache is passing users from web (click) into the messenger (bot). How we solved it:
- Pass the Click ID: When a user clicks an Adexium ad, we generate a unique
click_id. We append it to the bot link viastartapporstartparam. Example:t.me/our_bot?startapp={click_id} - Catch data in the bot: When the user hits
/start, our backend (Python) parses the param and ties it to their Telegram ID - Fire the postback: When the user completes a target action (buys premium or hits level 5 in-game), our server sends an S2S postback to Adexium with
price=10
The payoff: The network’s algorithms see which zones deliver paying users and automatically bump bids on them (Smart Bidding).
FAQ
Do I need pre-landers when driving traffic to Telegram?
Depends on the format. For In-Page Push, direct link to the bot works best; every extra step cuts conversion by 2–3x. For Popunder, a pre-lander is essential to “warm up” the user and filter out accidental clicks.
Does Telegram ban for arbitrage traffic?
Yes, if you’re pushing blatant spam or using bot farms. Adexium filters out incentivized traffic, so bot ban risk is minimal. Still, always warm up new bots before scaling hard.
What to take from this
Traffic buying for Telegram Mini Apps is one of the most profitable, and most technically demanding, markets right now. What works: click-worthy creatives + S2S postback on every action + daily zone pruning.
To try the same funnel, open an advertiser account. Popunder and in-page push carry most of our volume and convert into Telegram products when the pre-lander does its job. Wire the tracking, run a $200 test, and scale only the lines that hold their CPA.
