“Most organizations have bold strategies and real urgency to execute them.

But the current process to achieve the strategy is often inefficient and unprofitable, primarily due to shallow alignment, low accountability, and lack of focus on what matters most.

The OKR framework helps organizations tackle all three of these challenges.”

It surprised me because… well, that’s not what OKRs are for. At least not primarily.

OKRs and Long-Distance Running

A few years ago, I got into long-distance running. At first, it was just a way to stay fit, but soon I was training seriously for my first marathon. That’s when I got pulled into the world of training metrics: heart rate zones, VO₂ max, lactate thresholds, and so on.

What struck me over time was how powerful these metrics were—not just for tracking effort, but for learning. I could tell whether I was on track for my marathon goals without ever running a marathon. I didn’t need to wait until race day to find out. I had leading indicators. I had signals. I had information—and I had it early, and cheaply.

This, to me, is the true value of OKRs in business. Many people seem to think OKRs are a tool for accountability or performance tracking. But that misses the point. OKRs are a system for dialoguing with reality. They help you ask the world, “Is this working?”—and get answers early enough to matter. Here’s how this works

Reality Doesn’t Tell You the Truth for Free

In business, as in life, we act based on beliefs about how reality works. “If we work on Product A, we’ll increase revenue by 50% this year.” Maybe that’s true. But how do we know?

The most reliable way to find out would be to work on Product A, launch its new version, sell it hard for a year, and look at your revenue numbers. That’s the equivalent of running the full marathon. It’s also very expensive. If you were wrong—if the product doesn’t resonate, or it solves a problem customers don’t care about—you just sunk a year’s worth of time, people, and capital.

This is where OKRs come in.

OKRs Are Hypotheses About Reality

A good OKR isn’t about what you do. It’s about what you’re learning from the world. It’s not, “Build 80% of Product A’s roadmap this quarter.” That just tells you whether you can ship what you planned. It doesn’t tell you whether you shipped what you should have.

Instead, good key results to track would be:

• Product A active users grow by 30% in Q1

• User churn decreases by 10 percentage points next quarter

• 20% more deals close when Product A is pitched

• 50% of users have used Feature X at least once

These aren’t checkboxes for effort. They’re signals from reality. They help you see whether your actions are shaping the world the way you hoped and they do it early, so you can adapt without waiting until Q4 to find out you were wrong.

Think of OKRs as hypotheses. “We believe that if we do X, reality will respond in way Y.” Good OKRs let you test these hypotheses fast and affordably. That’s what makes them useful. It’s also why bad OKRs are worse than no OKRs at all—because they give you a false sense of progress. You can hit 100% of your goals and still be flying blind.

When companies have a bad quarter or a bad year, they often react by “doubling down on accountability.” That’s fine—but only if your OKRs are derived from reality, not from your internal plans. Otherwise, you’re just rewarding people for executing decisions that were wrong to begin with.

Most Organizations Confuse Output With Dialogue

In my experience, when things go wrong—missed targets, bad quarters—companies tend to double down on OKRs. “Let’s track harder. Let’s enforce more rigor.” But if the OKRs themselves are just internal checkpoints—“Did we build it?” “Did we ship it?”—then you’re just measuring your effort, not your impact.

You’re having a conversation, but only with yourself.

To be clear, measuring delivery can sometimes be useful. But if that’s all you’re doing, you’re not interrogating reality. You’re not learning. And that’s where predictability vanishes—where people are surprised by outcomes they should’ve seen coming months ago.

Running: A Model for Dialoguing with Reality

Here’s the analogy that made this click for me.

Suppose you want to run a marathon in under three hours. The most definitive way to find out whether you’re ready is… well, to run a marathon and see if you make it under three hours.

But that’s expensive. It takes months of training, weeks of tapering, days of carb loading, and a massive physical toll. If you fail, you’ve learned something—but you’ve paid a heavy price.

Instead, you watch the signals:

• What’s your pace at lactate threshold?

• Can you sustain race pace (4:15/km) without your heart rate crossing into Zone 5?

• Are you increasing the pace you can sustain at Zone 2?

These metrics tell you whether your physiology is adapting in the right direction. They let you predict race-day success with a high degree of confidence. More importantly, they let you adjust your training weeks or months in advance. You don’t just do threshold runs. You check after each one: Has my sustainable pace at threshold improved? That’s reality speaking. That’s information with a low price tag.

OKRs as an Ongoing Dialogue

So what are OKRs, really? They’re not to-do lists. They’re not rigid scorecards.

They’re structured conversations with the world.

Whether you’re training for a marathon or running a business, the question is the same: “If we do X, will Y happen?” And you’re trying to get an answer that’s honest, actionable, and—ideally—not too expensive.

If you’re using OKRs just to confirm your own intentions, you’re missing the point.

But if you use them to probe reality—to test your assumptions early and often—you’re giving yourself the ability to adjust before it’s too late.

That’s the difference between cheap insight and expensive illusion.

That’s what OKRs are for.