Case: 5,301 leads for $6,353 for a Telegram bot for the Aviator game

“Create a stable flow of deposits through the bot, grow the traffic budget without losing deposit volume, and hold the cost per lead to $1.5 for Kyrgyzstan and up to $2 for Azerbaijan.”

After 2–3 days of active traffic flow, leads started getting sharply more expensive and ad accounts started getting blocked.

We had to restart the ads from scratch every day, buying different types of ad accounts: farmed, auto-registered and agency accounts.

The combination was relatively simple: proxies → ad accounts → pre-lander (a transition link that opens the bot inside the app).

The main cause of blocks and rising traffic costs was the pre-landers and one-links: many links led to a single pre-lander, which raised suspicions at Facebook.

We solved the rising costs by refreshing creatives (including meta tags) and duplicating ads.

Communication with the client was a separate challenge: it required daily contact and explanations of how the ads worked, as the client did not understand why one day brought 100 leads and another — 87.

To solve this, we brought in and trained a junior media buyer who took over launching traffic and handling blocks, while the main communication with the client stayed with me.

The ad campaigns met and even beat the initial cost-per-lead requirements: the plan was up to $1.5 per lead for Kyrgyzstan and up to $2 for Azerbaijan, while the actual average cost per lead across the project was $1.1.

Working with ads and creatives

The main focus was on video creatives: they clearly showed the gameplay and examples of winnings, which converted into leads better.

In parallel we tested static creatives to find additional working combinations.

For scaling we used 3–4 active ad accounts, which let us grow the daily budget without a sharp drop in efficiency.

When traffic got more expensive, we regularly refreshed creatives and meta tags and duplicated ads to keep the cost per lead within the KPI.

We rebuilt the pre-landers and one-links to reduce the risk of blocks from Facebook.

Instead of chaotic launches and constant blocks, we built a more systematic way of working with accounts, creatives and pre-landers.

A clear scaling scheme appeared: several accounts, regular creative updates and cost-per-lead control for each country.

The client received a stable flow of leads at an average price of $1.1 against an initial requirement of $1.5–2.

Bringing in a media buyer reduced my operational load and the impact of the daily manual work with blocks on the results.

The project showed that in the scheme-traffic niche for an Aviator Telegram bot, results depend on the combination: geo → creative → pre-lander → accounts → daily optimisation. With the right setup we attracted 5,301 leads at an average cost of $1.1 and stayed within the client's cost-per-lead KPI.