Session
The Metrics Trap: When Measurement Distorts What It Aims to Improve
Organizations invest billions in data infrastructure and analytics, yet measurement systems often distort the very outcomes they aim to improve. This presentation reveals how metrics create structural dysfunction through four predictable patterns: proxy displacement (when indicators replace actual goals), feedback narrowing (when unmeasured dimensions atrophy), temporal compression (when short-term metrics override long-term viability), and strategic adaptation (when systems game metrics rather than improve performance).
Drawing from research across healthcare, education, and corporate sectors, I'll demonstrate how these patterns manifest in data-driven organizations. More importantly, I'll provide three implementable alternatives that maintain measurement's benefits while reducing distortion:
1. Feedback Diversification: Creating multi-channel feedback systems that capture qualitative signals alongside quantitative metrics
2. Adaptive Measurement: Building metrics that evolve in response to observed gaming and displacement
3. Structural Alignment: Separating measurement from immediate resource allocation to reduce optimization pressure
Attendees will learn to recognize early warning signs of metric dysfunction in their organizations and implement specific interventions at team, department, and enterprise levels. I'll share concrete examples from companies like Spotify, Adobe, and Kaiser Permanente that successfully transformed their measurement approaches.
Position measurement as a component within complex adaptive systems rather than the system itself. Attendees will leave with a practical framework for building measurement systems that enhance rather than distort organizational performance.
Key Takeaways
* Diagnostic Framework: Learn to identify the four patterns of metric dysfunction before they degrade system performance
* Implementation Toolkit: Gain specific techniques for feedback diversification, adaptive measurement, and structural alignment
* Transition Strategies: Understand how to overcome institutional dependencies, technical limitations, and cultural resistance when transforming measurement approaches
* Cross-Scale Application: Apply solutions from individual contributor level through enterprise-wide systems

Darryl Kanouse
Head of Product & AI/ML, Data Science & Engineering @ Amazon Web Services (AWS) | Viability Architect @ The Autogenic Realist | Writer
Redondo Beach, California, United States
Links
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top