> Markdown version of [/videos/100317-measuring-the-wrong-things-faster-than-ever?t=1373](https://www.wearedevelopers.com/videos/100317-measuring-the-wrong-things-faster-than-ever?t=1373). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Measuring the Wrong Things Faster Than Ever

AI tools write 17x more code, yet releases only grew 1.3x. Stop optimizing for dangerous token leaderboards. Discover how to measure true engineering outcomes over AI-generated noise.

- **Speakers:** [Laura Tacho](https://www.wearedevelopers.com/@laura-tacho)
- **Event:** World Congress 2026 Europe
- **Published:** July 10, 2026
- **Duration:** 30:47
- **URL:** https://www.wearedevelopers.com/videos/100317-measuring-the-wrong-things-faster-than-ever

## Summary

Engineering teams are rapidly adopting AI-driven coding tools, reportedly writing 17 times more code but only seeing a minor 1.3x increase in actual software releases. As AI operational costs skyrocket, engineering leaders face mounting pressure to prove ROI and frequently fall back on easy, "one-dimensional" metrics like token leaderboards or AI-generated lines of code. However, just as dominating ball possession in a soccer match doesn't guarantee a win, optimizing for token output provides a dangerous illusion of productivity that ultimately falls victim to Goodhart's law: once a metric becomes a target, it ceases to be a good metric.

The surge in AI-generated code has introduced severe second-order effects across the software development life cycle (SDLC). Current industry data reveals that code churn has spiked by over 800%, and median pull request sizes have nearly doubled, crippling code review pipelines. Consequently, teams are experiencing "code review surrender"—where overwhelming PR volume leads to superficial checks or skipped reviews—and "code review exploitation," where reviewers feel unfairly burdened with evaluating unrefined, agent-generated code. These bottlenecks significantly erode team trust and polarize overall deployment quality.

To avoid measuring the wrong things faster than ever, organizations must pivot from "token maxing" to "outcome maxing." Effective measurement requires a robust evaluation framework that strings multiple metrics together to create natural tension. Engineering leaders should use the word "without" to frame goals—such as increasing throughput *without* raising review wait times or degrading developer experience. By tying metrics to explicit organizational outcomes, analyzing data at the team level instead of the individual level, and continuously benchmarking against past performance, companies can accurately capture AI's true return on investment.

**Keywords:** developer productivity metrics, token leaderboards, generative AI ROI, code review surrender, code review exploitation, code churn rates, SDLC bottlenecks, pull request bottlenecks, autonomous coding agents, software engineering intelligence, AI-generated code quality, engineering team performance, outcome maxing, metrics gamification, Goodhart's law

## Chapters

1. **The illusion of dominance in one-dimensional metrics** (00:02) — Comparing actual match outcomes to ball possession time reveals how one-dimensional measurements fail to accurately predict success.
1. **Anatomy of misleading metrics and token leaderboards** (05:14) — Token leaderboards and proxy metrics fall victim to Goodhart's law when organizations track them without proper context.
1. **Industry saturation and the rise of autonomous agents** (08:26) — As initial tool adoption hits saturation, the shift toward autonomous agents fundamentally impacts enterprise software development pipelines.
1. **The widening gap between code authored and released** (12:21) — While artificial intelligence generates more total volume, surging pull request sizes and severe code churn throttle actual release rates.
1. **Addressing code review surrender and process exploitation** (14:55) — A massive influx of generated modifications leads to capacity failures and trust erosion during decentralized review processes.
1. **Divergent trends in production failures and change confidence** (18:34) — Average quality measurements fail to capture a polarized reality where some organizations halve operational failures while others double them.
1. **Skyrocketing tool spending and improving token efficiency** (20:05) — Rapidly growing monthly platform costs per developer drive a deeper industry focus on model efficiency and open weights.
1. **Principles for balancing multiple software engineering health metrics** (22:53) — Tying organizational decisions to competing systemic checks ensures throughput gains do not unintentionally degrade code quality or developer experience.
1. **Shifting leadership focus from token maxing to outcome maxing** (27:53) — Replacing isolated leaderboards with unified strategic frameworks accurately aligns technological investments to critical business and developer outcomes.

## Related Moments

- [Measuring software impact instead of lines of code](https://www.wearedevelopers.com/videos/100106-craftsmanship-in-the-age-of-agents) (from "Craftsmanship in the Age of Agents")
- [Measuring developer productivity, efficiency metrics, and team happiness](https://www.wearedevelopers.com/videos/100256-can-this-elephant-dance-ibm-bob-and-the-future-of-ai-first-software-development) (from "Can This Elephant Dance? IBM Bob and the Future of AI-First Software Development")
- [Measuring developer productivity without killing engineering flow](https://www.wearedevelopers.com/videos/1691-engineering-productivity-cutting-through-the-ai-noise) (from "Engineering Productivity: Cutting Through the AI Noise")
- [Motivations for adopting AI to enhance developer productivity](https://www.wearedevelopers.com/videos/1266-navigating-the-ai-revolution-in-software-development) (from "Navigating the AI Revolution in Software Development")
- [Overcoming review fatigue and measuring true engineering productivity](https://www.wearedevelopers.com/videos/1706-the-ai-ready-stack-rethinking-the-engineering-org-of-the-future) (from "The AI-Ready Stack: Rethinking the Engineering Org of the Future")
- [Measuring generative AI impact on team productivity](https://www.wearedevelopers.com/videos/100026-building-10x-organizations-using-modern-productivity-metrics) (from "Building 10x Organizations Using Modern Productivity Metrics")

## Related Articles

- [Transforming Software Development: The Role of AI and Developer Tools](https://www.wearedevelopers.com/magazine/527-transforming-software-development-the-role-of-ai-and-developer-tools)
- [One billion (bad?) developers: How AI is changing the way we learn to code](https://www.wearedevelopers.com/magazine/516-one-billion-bad-developers-how-ai-is-changing-the-way-we-learn-to-code)
- [What is Software Engineering in the Age of AI?](https://www.wearedevelopers.com/magazine/640-what-is-software-engineering-in-the-age-of-ai)
- [Exploring AI: Opportunities and Risks for Developers](https://www.wearedevelopers.com/magazine/522-exploring-ai-opportunities-and-risks-for-developers)

## Related Jobs

- [Staff Software Engineer, Copilot Experiences](https://www.wearedevelopers.com/jobs/ext/164361-staff-software-engineer-copilot-experiences) at **GitHub**
- [Tribe Lead - ( Software) Engineering Centre of Excllence](https://www.wearedevelopers.com/jobs/ext/1475530-tribe-lead-software-engineering-centre-of-excllence) at **SD Worx**
- [Principal Software Engineer, Enterprise AI Platform](https://www.wearedevelopers.com/jobs/ext/1467292-principal-software-engineer-enterprise-ai-platform) at **GitHub**
- [Senior AI Agent Software Engineer (Go, Python) (m/f/x)](https://www.wearedevelopers.com/jobs/48277-senior-ai-agent-software-engineer-go-python-m-f-x) at **Dynatrace**
- [Principal Product Manager, Agent Platform](https://www.wearedevelopers.com/jobs/ext/277541-principal-product-manager-agent-platform) at **GitHub**
- [Staff Developer Advocate, GitHub Security Lab](https://www.wearedevelopers.com/jobs/ext/1921051-staff-developer-advocate-github-security-lab) at **GitHub**