---
id: "contrarian-productivity-vs-capability"
type: "contrarian-insight"
source_timestamps: ["§ Three Necessities", "¶16", "¶17"]
tags: ["contrarian-insight", "metrics", "ai-collaboration", "evaluation"]
related: ["action-analyze-task-level", "contrarian-skills-based-obsolescence"]
speakers: ["Sangeet Paul Choudary", "John Winsor"]
challenges: "The assumption that higher output volume via AI tools directly equates to higher employee capability or value."
sources: ["tail1"]
sourceVaultSlug: "hbr-seg-tail1"
originDay: 1
articleStem: "hbr-tail-112-continually-assessing-performance"
sourceUrl: "https://hbr.org/2026/06/the-pros-and-cons-of-continually-assessing-performance"
sourceTitle: "The Pros and Cons of Continually Assessing Performance"
---
# Contrarian — Productivity Measurement Is Not Capability Assessment in the AI Era

**Contrarian insight.** *Challenges:* the assumption that higher output volume via AI tools directly equates to higher employee capability or value.

Traditional productivity measurement focuses on **output volume**. The authors argue that in the AI era, capability assessment is a fundamentally different thing: it measures *how well a human supervises, evaluates, and integrates AI-generated output into reliable systems* — not the raw amount of output produced.

An employee who uses AI merely for *acceleration* is demonstrating a different (and potentially less valuable) capability than one who *deconstructs and manages* AI workflows. This reframes what the task-level signals in [[action-analyze-task-level]] should be read for: not "who ships more," but "who excels at supervising AI output." It reinforces the broader thesis that a static skills catalogue misreads value (see [[contrarian-skills-based-obsolescence]]).

**Counter-perspective (from enrichment):** telemetry does not automatically equal capability. Tool-usage metrics, acceptance rates, and communication traces reveal *activity* but can miss judgment, creativity, collaboration quality, and long-horizon problem-solving — which is exactly why the article calls for contextual interpretation (see [[claim-contextual-performance-variation]]).