Boxes must be Green, Lines must go Up
“The moment you make a metric a target, it ceases to be a good metric.” — Charles Goodhart, economist
Picture this: It’s another quarterly planning session at your typical tech company. A young engineering manager sits across from his team, laser-focused on their traffic migration goals. The energy is palpable — everyone wants those green boxes on the quarterly slides. After all, nothing says success quite like an upward-trending line on a chart, right?
Well, let me tell you a story that might make you think twice about that assumption.
The JavaScript File That Broke the System (And a Manager’s Perspective)
Our young manager was riding high. His team had identified a JavaScript file receiving significant traffic in the legacy system they were tasked with deprecating. The numbers were beautiful — migrating this single static file would represent a massive chunk of their quarterly traffic migration target. It was almost too good to be true: minimal effort, maximum impact on the KPI.
They prioritized it immediately. Why wouldn’t they? The migration was straightforward, the impact substantial, and the path to those coveted green boxes crystal clear.
A few weeks later, during the biweekly stakeholder check-in with Directors and Principal Engineers, our manager proudly presented the dramatic jump in progress. The charts looked fantastic. Then came the question that would haunt him: “Shouldn’t JavaScript files be in S3 and served via CDN static content domains?”
His response? “That’s not part of my goals — refactoring should be a different effort.”
And there it was. The first mistake. Blinded by the beauty of his metrics, he’d forgotten to ask the most important question: What’s the right thing to do?
Moving a JavaScript file to the new system versus moving it to proper static hosting — the effort difference was negligible. But this option wasn’t even considered because it didn’t serve the number. As Einstein once said, “Not everything that can be counted counts, and not everything that counts can be counted.” Our manager had fallen into the trap of letting what could be counted override what actually counted.
Here’s the kicker: the KPI itself might have been fundamentally flawed. When you’re deprecating a system, shouldn’t you measure traffic reduction from the old system rather than just migration volume? But questioning the metric wasn’t on anyone’s radar when the line was trending so beautifully upward.
The Framework Migration That Taught Empathy
I’ve seen this pattern countless times, and here’s another story that illustrates just how seductive those green boxes can be.
I once tasked a manager with coordinating a framework migration across multiple teams in our area. He returned triumphant, armed with commitments from every team to complete their portion by next quarter. For him, this was the ultimate win — framework migration completed in record time, boxes turning green across the board.
The reality? He’d secured these commitments without considering what else was on each team’s plate. Some teams had more critical priorities that he simply hadn’t asked about. As quarters progressed, most teams delayed the framework migration as their real priorities took precedence, leaving our manager frustrated and blaming the teams for not honoring their commitments.
The truth was more nuanced. He’d been part of the problem, focusing solely on his goal without empathy for the broader context each team was operating within.
As Oscar Wilde wryly noted, “Experience is merely the name men gave to their mistakes.” The question is: are we creating environments where people can accumulate the right kind of experience without fear of retribution?
The Winning Trap: When Success Becomes the Enemy
We’re culturally wired to win. We grow up watching sports where heroes get gold medals and trophies. That competitive drive serves us well in many contexts, but in engineering leadership, it can become a dangerous blind spot.
Business requires objectivity. It demands that we not only acknowledge failure but be comfortable with it. Nobody wants to fail, of course, but you absolutely need to if you want to grow and improve. The measure of a great engineering leader isn’t found in their successes — winning is the happy path. It’s in how they handle failure, learn from it, and adapt.
This is where Extreme Programming got it right with their embrace of failure as a fundamental learning mechanism. XP practices like test-driven development and continuous integration aren’t just technical disciplines — they’re philosophical statements about the value of rapid feedback and learning from mistakes before they become catastrophic.
When we create cultures obsessed with green boxes and upward lines, we inadvertently punish the very experimentation and honest assessment that drives real innovation and improvement. We optimize for the appearance of success rather than actual progress.
As Eric Ries warns in The Lean Startup: “Vanity metrics are the numbers you want to publish on TechCrunch to make your competitors feel bad.” This is exactly what we’re doing, and its not how you drive the business.
Beyond the Dashboard: What Really Matters
The irony is that our obsession with metrics often prevents us from achieving what those metrics were supposed to measure in the first place. When you focus exclusively on migration traffic without considering architectural best practices, you might hit your target while actively making the system worse. When you coordinate commitments without understanding competing priorities, you might get green checkmarks while destroying trust and collaboration.
As management consultant Peter Drucker famously said, “Efficiency is doing things right; effectiveness is doing the right things.” We’ve become incredibly efficient at moving metrics, but have we lost sight of effectiveness?
Your achievement as an engineering leader extends far beyond a green box or an ascending line on a chart. You’re building systems, nurturing teams, and solving real problems for real users. Don’t let those broader impacts get lost in the pursuit of dashboard aesthetics.
A Way Forward: Three Principles for Metric-Conscious Leadership
After years of watching talented managers fall into these traps (and occasionally tumbling into them myself), I’ve learned that sustainable success requires a different approach:
1. Embrace failure as a learning experience. It’s not just okay to fail — it’s necessary. Create space for your team to experiment, to question assumptions, and yes, to occasionally miss targets if it means learning something valuable. The goal isn’t perfection; it’s continuous improvement.
2. Focus on your goal, not the number. If the metric isn’t accurately reflecting progress toward your actual objective, it’s okay to suggest changing how you calculate it. Numbers should serve the mission, not the other way around. Be brave enough to question whether what you’re measuring actually matters, and even to suggest a course of action that might not impact it or even make it worse if its the right decision, because maybe we are measuring the wrong thing.
3. Think holistically and practice empathy. When working with other teams, understand their context, their competing priorities, and their constraints. Your green box shouldn’t come at the expense of someone else’s critical work. True success is achieved when the we do the right thing for the company, not for the individual or the team.
Remember: you’re not just a steward of metrics — you’re a steward of people, systems, and progress. Those green boxes and upward lines should reflect genuine value creation, not just clever number manipulation.
The best engineering leaders I know have learned to love the messy complexity of real progress over the clean simplicity of dashboard success. They understand that sometimes the most important work doesn’t immediately show up in the metrics, or maybe its even something that can’t be measured, and that’s perfectly fine.
After all, as the saying goes, “The map is not the territory.” Your metrics are just the map — don’t mistake them for the real terrain you’re trying to navigate.