Lasta

Lasta is a weight loss app that leverages web-based funnels to acquire and convert users into paid subscribers. The product operates through multiple funnel variations and actively scales acquisition using AI-generated UGC creatives and performance-driven funnel logic.

For this campaign, the focus was on building a predictable and scalable paid acquisition system via NewsBreak, targeting users aged 40+ interested in weight loss. The objective was to drive high-quality paid users within a controlled CAC range of $60–$72, while maintaining stability across fluctuating auction conditions.

Before scaling on NewsBreak, several key challenges limited growth:

• Creative fatigue, causing rapid performance drops
• Unstable performance at scale, especially with increasing budgets
• Fluctuating auction dynamics, impacting CPA consistency
• Difficulty maintaining cost control while scaling paid users
• Limited ability to react quickly to performance changes without automation

We built a performance-driven acquisition system on NewsBreak, combining structured funnel logic, AI-powered creatives, and automation.

Funnel Structure

NewsBreak Ad → Pre-lander → Registration (Lead) → Purchase

• Purchase event tracked via S2S integration (100% attribution accuracy)
• Upper funnel events tracked via pixel for optimization signals
• Focus on downstream purchase performance

Campaign Setup & Targeting

• Target CPA bidding strategy for stable cost control
• Campaigns split by:
• Gender (male/female)
• Age targeting (40+)
• Separate ad sets for each segment to improve optimization precision

Creative System

• AI-generated UGC vertical videos (~60 sec)
• Storytelling approach:
• Real-life weight loss scenarios
• Urgency-driven narratives
• AI characters in diverse environments

Execution:

• ~50 new creatives launched weekly
• Continuous testing across multiple funnels
• Creative aligned with funnel logic for better conversion quality

Automation & Optimization (Point2Web)

To eliminate manual inefficiencies, a custom automation system was implemented:

Autorules included:

• Auto-pause underperforming ads (works both in early learning and high-spend stages)
• Auto-reactivation of ads (re-enables ads that regain performance)
• Automated creative rotation (injects fresh creatives into selected ad sets)

Advanced logic:

• Different rules based on conversion volume
• Flexible AND/OR conditions
• Multi-metric optimization decisions

Fully automated lifecycle:
Detect fatigue → Pause → Replace → Relaunch

Optimization Approach

• Focus not only on CPA but also on conversion rate (CR)
• Identification of real performance drivers vs. auction fluctuations
• Continuous monitoring to avoid misinterpreting competition spikes as creative fatigue