LIVE: variable stacking. (the secret to scaling your ads fast af)

Mark introduced 'variable stacking' as a framework for scaling ads profitably by treating ad optimization through the scientific method lens of independent and dependent variables. Rather than testing multiple variables simultaneously and hoping for winners, the strategy involves isolating and testing each variable category separately, identifying the winner in each category, then stacking all winners together for maximum performance.

The basic variables Mark covers are creative (specifically creative format as a 'quantum leap'), copy (especially important for static/image ads), and funnel type (advertorial, listical, straight-to-PDP, sales page). Once winners are identified in each category, they're combined for baseline scaling.

Marк then explores advanced variables beyond the core three. Payment processors significantly impact profitability through approval rates and AOV—testing multiple processors with identical winning creatives can reveal 30-50% differences in approval rates. Facebook ad infrastructure variables include ad accounts and pages; some ad accounts inherently outperform others due to reputation and data history with Facebook, making this a worthwhile test. Similarly, page types (brand, personality-based, general, whitelisted) show measurable performance differences.

Website hosting platforms represent another critical variable—Mark tested Shopify, Replo, ClickFunnels, Funnelish, and Checkout Champ with identical pages and found Shopify typically wins on conversion rate, though other platforms occasionally outperform on profitability. AI copywriting sources can be tested against each other by win rate (how many ads become profitable). Landing page builders and custom-coded AI-generated pages also qualify as testable variables.

Media buying structure variables include campaign setup types (CBO with bid caps vs. manual bidding), audience targeting (demographics and placements), and creative testing environments designed to give new creatives the highest probability of success by surrounding them with all other proven variables.

Mark emphasizes that scaling is fundamentally a game of testing extensively, finding winners in each variable category, then directing budget toward the combined stack of all winners rather than spreading spend across unproven variables. He contrasts this with most people who either don't test variables systematically or test new variables in isolation without stacking them with existing winners.