---
id: "cp-data-infrastructure-bottleneck"
type: "counter-perspective"
source_timestamps: ["Enrichment: Counter-Perspectives §1", "Enrichment: Claim Validations"]
tags: ["counter-perspective", "data-infrastructure", "telemetry", "bottleneck"]
related: ["claim-bottleneck-is-explicit-judgment", "claim-deployment-is-table-stakes", "concept-judgment-infrastructure"]
sources: ["agentic"]
sourceVaultSlug: "hbr-seg-agentic"
originDay: 6
articleStem: "hbr-new-27-teach-ai-your-decisions"
sourceUrl: "https://hbr.org/2026/06/teach-your-ai-how-you-make-decisions"
sourceTitle: "Teach Your AI How You Make Decisions"
---
# Counter: The Primary Bottleneck Is Data & Telemetry, Not Judgment

**Competing thesis (from HBR Analytic Services + Cribl):** The biggest barrier to agentic-AI success is fragmented and unreliable **data** and legacy observability/security stacks — not judgment codification. The organizations pulling ahead are re-architecting their data layer, treating telemetry as strategic input, and building open, interoperable platforms.

**Why it challenges the source:** It relocates both the bottleneck ([[claim-bottleneck-is-explicit-judgment]]) and the moat ([[claim-deployment-is-table-stakes]]) from judgment to data/telemetry infrastructure.

**Implication:** [[concept-judgment-infrastructure|Judgment infrastructure]] may be ineffective if built on poor data foundations; some experts argue data readiness must precede or at least match judgment codification. Even well-codified judgment can be misapplied without reliable telemetry to feed it. This is complementary rather than strictly contradictory — both can be true simultaneously.