---
title: "Own the Asset: Proprietary Data and Structural Moats"
arc: "competitive-advantage"
articles: ["a107", "a109", "a116"]
tags: ["cross-day", "moat", "build-vs-buy", "data-as-asset"]
id: "cross-proprietary-asset-moat"
sources: ["tail1"]
type: "synthesis"
sourceVaultSlug: "hbr-seg-tail1"
originDay: 1
articleStem: "hbr-seg-tail1"
sourceUrl: "(unified vault: 14 sources)"
sourceTitle: "HBR — Tail Ⅰ · Adjacent — firm, people, demand, futures (#104–117)"
---
Three articles treat *ownership of a hard-to-replicate asset* as the durable source of advantage — but disagree on how absolute that should be.

- **A107**: [[concept-proprietary-operational-data-advantage|proprietary operational data is an un-replicable moat]]; hence [[contrarian-build-vs-buy-ai|build the AI architecture internally]] rather than buying off-the-shelf ([[claim-off-the-shelf-ai-inadequate]]).
- **A109**: data is *so* valuable that it should be *priced and paid for* — [[concept-data-mixture-weights|mixture weights]] and [[concept-scaling-laws-valuation|scaling laws]] estimate its share ([[claim-data-value-percentage]]), and [[claim-data-exhaustion|the free stock is running out]], making fresh data an investment, not a tax.
- **A116**: the moat can be *organizational* — [[concept-structural-separation-commitment|legally separating a venture]] (Microsoft/OpenAI) protects a commitment position that generic capital cannot buy.

**The productive disagreement:** A107 preaches full internal build; the enrichment across the corpus warns this is the *most contested* bet — hybrid "compose-and-customize" often wins, and A109 shows platforms can license the underlying asset instead. The synthesis: the *asset* (proprietary data, committed structure) is the moat; whether you *build, buy, or license the plumbing around it* is a scale-dependent choice. Cross-link: [[cross-data-foundation-prerequisite]].