Modeled conversions inflate reported results by 9–17% Across 26 accounts cross-checked vs backend, 90d. Modeled = conversions Meta estimates, not observed. — Share of reported conversions that are mod…
Thumbstop rate under 30% is the real fatigue signal — before CPA moves Across 40 accounts, 90d. Thumbstop = 3-sec video views per impression. — Top quartile creative: 42%+ — Median: 32% — Bottom:…
Your breakeven ROAS is usually 30% higher than you think Across 35 DTC P&Ls reviewed, 90d. Breakeven ROAS = 1 / gross margin. — 40% margin product: breakeven ROAS 2.5x — Add 6% payment+shipping: real …
Front-loaded daily budgets spend 60% by 2pm and pay for it Across 29 accounts with intraday tracking, 90d. — Morning-heavy pacing (no schedule): CPM ▲11% vs evening hours — Even pacing via dayparting:…
Cost cap delivers 14% lower CPA than lowest-cost — at 30% less volume Across 24 accounts A/B'd on the same campaign, 60d. — Lowest-cost bid: median CPA $41, full volume — Cost cap (set at target): CPA…
A high CTR with low CVR means a creative problem, not an audience one Across 44 underperforming ad sets, diagnosed by metric split. CVR = conversions per click. — Low CTR + normal CVR: weak hook/creat…
