A crash-game iGaming offer (RevShare model) run into Peru via Facebook and Telegram over three weeks — first market test for the team, closing at 145% ROI on a modest initial budget of $4,100.
The task
The advertiser came to us with a funnel that already worked but had hit a volume ceiling. We needed to run a first test of the crash game on a RevShare model in the Peruvian market and see whether the approach that worked earlier in Brazil would repeat.
We worked the "Crash games" vertical in the Peru market. Payment model — RevShare, main sources — Facebook Ads, Telegram bot.
How we did it
The setup was built sequentially, testing hypotheses on small budgets.
- Launched a player-controlled cash-out mechanic — it led to longer, more emotionally engaged sessions than standard slot content.
- Connected a Telegram bot pre-qualifying and "warming up" traffic before it reached the offer — modeled on the approach that had worked well in the earlier Brazilian crash campaign.
- Tested a direct Facebook → offer flow in parallel with the bot flow to compare conversion.
- Made Spanish-language creatives with Peruvian slang instead of generic Latin American Spanish.
What went wrong
Not everything went according to plan. The Peruvian market had not been tested before, and it was an open question whether the same mechanic that worked in Brazil would work there, or whether local differences in competition and language would require a separate approach.
Conversion on traffic warmed by the bot turned out to be roughly double the control group with the direct flow, and Spanish creatives with Peruvian slang delivered a noticeable lift over neutral text — both tactics confirmed a pattern already seen in other Latin American markets.
The result
Here are the final numbers for the buying period. Over the three weeks: roi — 145%, spend — $4,100.
The campaign closed at 145% ROI on a starting budget of $4,100 — CPC came in noticeably lower than in comparable Brazilian campaigns, reflecting lower competition for this vertical in the Peruvian market.
What we took away
- Warming up traffic through a Telegram bot before it reaches the offer consistently lifts conversion in crash games.
- Genuinely regional language work (local slang) beats broad regional generalization on almost every Latin American market tested.
These numbers belong to one specific setup, geo, and buying period. Economics will differ on another offer or in another season — we always recalculate them before launch.