Netflix's default practice of validating major, risky, or uncertain product changes — ad banners, plan-wide ad tests, pricing changes — through large-scale A/B experiments before wide rollout, rather than shipping on conviction alone. Biddle notes one exception: when a competitor has already proven the model in market, testing can be skipped since the market has effectively already run the experiment. Sits downstream of DHM Framework (Delight, Hard-to-Copy, Margin-Enhancing) and Two-Question New-Initiative Filter in Biddle's Strategy → Plan → Consumer Science → Product Sense Pipeline — the 'consumer science' step that can resolve a question before product sense (a SWAG (Stupid/Scientific Wild-Ass Guess)) is needed.
Netflix skipped big A/B testing before launching advertising both times it did so — the original ~2005-era ad business ('they just did it') and the 2022 ad-tier relaunch, where Hulu's ad tier had already validated demand, making a dedicated internal test feel unnecessary. This suggests the 'test everything' default has a carve-out: when a close competitor has already run the real-world experiment, pattern-matching off their results can substitute for an internal A/B test.
Из тем: Netflix, Consumer Science, and Strategy Storytelling