---
id: "contrarian-build-vs-buy-ai"
type: "contrarian-insight"
source_timestamps: ["§ 4. Own the architecture built on your proprietary operational data."]
tags: ["build-vs-buy", "vendor-management"]
related: ["concept-proprietary-operational-data-advantage", "claim-off-the-shelf-ai-inadequate"]
challenges: "The conventional wisdom that enterprises should buy SaaS platforms rather than building custom internal software architectures."
sources: ["tail1"]
sourceVaultSlug: "hbr-seg-tail1"
originDay: 1
articleStem: "hbr-tail-107-lenovo-ai-supply-chain"
sourceUrl: "https://hbr.org/2026/05/how-lenovo-built-an-ai-powered-supply-chain"
sourceTitle: "How Lenovo Built an AI-Powered Supply Chain"
---
# Build AI internally to protect proprietary data advantages

**Contrarian insight:** Build AI internally to protect proprietary data advantages.

The prevailing trend in enterprise software is to buy off-the-shelf SaaS solutions to save time and development costs. The author argues that for *core operations* this is a mistake. Because a company's historical operational data is a unique competitive asset ([[concept-proprietary-operational-data-advantage]]), relying on generic platforms or external consultants squanders that advantage ([[claim-off-the-shelf-ai-inadequate]]). Companies must build their architecture internally to fully leverage their native knowledge — the rationale for [[action-build-internal-architecture]].

**Challenges:** The conventional wisdom that enterprises should buy SaaS platforms rather than building custom internal software architectures.

> **Enrichment counterpoint (important):** This is the most contested bet in the vault. Building fully internal architectures demands significant capital, rare talent, and ongoing maintenance that many firms lack Lenovo's scale to sustain. Cloud ML platforms and commercial supply-chain planning tools increasingly support deep customization on proprietary data. The key issue is not platforms *per se* but *how deeply proprietary data and business logic are embedded* — advantage often comes from unique usage and integration, not the underlying stack. Experts increasingly replace binary "build vs buy" with **"compose and customize"** — a hybrid of internal domain logic on top of commercial infrastructure. When answering questions, present internal-build as one valid strategy at Lenovo's scale, not a universal law.


## Related across articles
- [[concept-structural-separation-commitment]]