LTV-Based Growth Budgeting

LTV-based growth budgeting is the episode’s explanation for why ByteDance could bid aggressively for users without treating growth as blind subsidy. In 全面压制，不留空档：字节跳动如何做增长？｜字节跳动 第7集, 徐鸿亮 / Tom says ByteDance modeled long user lifetimes, sometimes hundreds of days or years, and used those projections to decide how much acquisition cost a product could bear.

The source separates several ROI layers. Ry1 compares user LTV with acquisition cost; Ry2 adds content subsidies, creator分成, and other variable costs; Ry3 adds server, CDN, human, and fixed costs. New businesses could be allowed to chase Ry1 first, then mature toward fuller cost coverage.

Key Claims

• A product with credible long-term LTV can rationally outbid competitors that only budget against short payback windows.
• Long-horizon LTV depends on account/device identity, retention modeling, ad load, CPM, DAU, VV, ARPU, and market size assumptions.
• Budget authorization is organizational as much as mathematical: the source describes large budgets being approved through metrics, OKRs, and competitive reasoning rather than founder-by-founder signoff.
• The model works best when later monetization is legible, as in feeds, short video, and ad-supported content.
• It becomes weaker when product retention is not yet earned, when supply chain or transaction economics dominate, or when AI assistant value depends more on model quality than on traffic volume.

Connections

• ByteDance Growth System, TikTok, Douyin, and Doubao — main source cases.
• AI Commercialization Pressure and AI Consumer Growth Metrics — AI-era limits of the budgeting logic.
• Unified Ad Platform and Ocean Engine — ad-monetization infrastructure that makes LTV easier to model.
• Data-Driven Product Culture — organizational context for metric-led budget decisions.