Bidding and Algorithm Control
Value-based bidding architecture: designing conversion value schemes using margin, predicted LTV, or lead-stage values, and applying conversion value rules (by location, device, audience) to steer tROAS toward profit rather than revenue.
Marginal vs. average ROAS thinking: using Performance Planner, bid strategy simulators, and spend-response curves to find where the next dollar stops being profitable, rather than optimizing to a blended ROAS target.
Learning-phase management: knowing what resets learning, when to use seasonality adjustments for short predictable spikes (sales, launches), and when to apply data exclusions after tracking outages.
Portfolio bid strategies with max/min CPC bid limits, shared budgets, and target impression share used deliberately for brand defense.
New customer acquisition goals: configuring “bid higher for new customers” or “new customers only” modes using Customer Match lists and conversion-based customer detection.
