Stop Measuring Only Outputs. Start Valuing the Process.

TL;DR: In an AI-native world, it's too easy to get bamboozled with seemingly good outputs. This creates shallow, brittle work and organizations. The real competitive advantage comes from leaders who review the system thinking that led to those outputs: the process, decomposition, and learning agility that actually drive results.

Apha School and everything wrong with output-focused education

I was listening to a Hard Fork episode (highly recommend this New York Times' podcast) featuring the founder of Alpha School, a new educational approach powered by AI promising "guaranteed outcomes." The pitch was ambitious: students overachieve and learn faster than traditional education.

But something felt deeply wrong about this framing.

My problem isn't with the goal of accelerated learning. It's with what we're optimizing for. Alpha School's approach mirrors a dangerous pattern I see across corporate America: an obsession with outputs at the expense of the process that creates them.

When we focus solely on outcomes (test scores, polished campaigns, final deliverables), we miss what actually builds lasting value. The real objective isn't accumulating knowledge or producing outputs. It's learning how to learn, how to decompose problems, and how to think systematically. And we'll skip the social elements for now.

This isn't just an education problem. It's a management problem.

The corporate parallel: output obsession at work

The same pattern plays out in professional settings every day. I've watched marketing teams use ChatGPT to generate final assets with no scaffolding, no audience definition, no strategic thinking. They get a deliverable, but it's brittle. When the context changes or feedback comes in, they have no foundation to iterate from.

As a result they produce shallow work that looks polished but lacks depth.

I've seen this across functions. Engineering teams start with AI as autocomplete, churning out code that "works" but doesn't scale. Product teams use AI to generate user stories without understanding the underlying user journey. Sales teams create outreach templates without mapping the buyer's decision process.

In each case, the focus on output creates a dependency trap. Teams become good at producing deliverables but terrible at the thinking that makes those deliverables valuable.

What happens when process becomes the focus

Here's what changed when I shifted my team's approach from output-focused to process-oriented thinking.

Relevant and nimble marketing team

I instructed our marketing team to stop asking ChatGPT for final deliverables. Instead, I had them:

1. Define the audience first - Who are we talking to? What do they care about? What's their current mental model?
2. Break the message into components - What's the core insight? What's the supporting evidence? What's the call to action?
3. Use ChatGPT step-by-step - Prompt for help within each component, not for the final product

The results were immediate and dramatic. Content quality improved 10x. Internal stakeholders gave clearer feedback because they could see the thinking behind each piece. The team developed reusable frameworks instead of one-off outputs.

Highly efficient engineering team

Our engineering team followed a similar evolution. They started with Cursor as autocomplete, generating code that worked but wasn't truly transformative. As we matured our AI practice, they shifted toward building systems, reusable logic, and better workflows.

The breakthrough came when they stopped asking "Can AI write this function?" and started asking "How do the constraints impact the architecture?" The output improved because the process put focus on the right things. The drudgery of writing code could then be effectively delegated to AI because they had thought through the problem and the constraints.

The process matters more than the outcome

This isn't just a management theory. It's something I learned the hard way.

In undergrad, I learnt to appreciate the art of mathematical proof. Our educational system gave zero credit for correct answers if the proof was not provided. At first, it felt overly zealous. I quickly realized I was being taught something more valuable than math.

The process matters more than the outcome.

Getting the right answer isn't enough. You need to show the path. You need to demonstrate the thinking that led to the solution. You need to make your reasoning visible so others can follow it, critique it, and build on it. I often found that what seemed trivial actually might have some corner cases that I missed.

This applies directly to how I manage teams today. When we only evaluate outputs, we're essentially giving credit for the right answer without caring about the work that produced it. We're rewarding luck over skill, surface over substance.

Practical tips: three questions that change everything

The biggest change managers need to make isn't in their tools or processes. It's in the questions they ask.

Instead of "Are we on track to deliver our project?" or "When will this be done?", start asking:

1. "How did you arrive at this?"

This question forces people to articulate their reasoning. It reveals whether they're thinking systematically or just producing outputs. You'll quickly discover who's building on solid foundations versus who's winging it.

2. "What thinking went into it?"

This goes deeper than the first question. It's about understanding the mental models, assumptions, and frameworks someone used. It helps you evaluate the quality of their thinking, not just the quality of their output.

3. "Can others follow your process?"

This is the ultimate test. If someone can't explain their process clearly enough for others to replicate it, they're not building organizational capability. They're just producing individual outputs.

There's a French saying: "ce qui se conçoit bien s'énonce clairement" which translates to "what is well understood is well expressed".

These questions do more than improve evaluation. They change behavior. When people know they'll be asked to explain their thinking, they start thinking more systematically. They build better frameworks. They create reusable processes instead of one-off solutions.

Process as competitive advantage in an AI-native world

This isn't just about better management. It's about building organizations that can thrive in an AI-native world.

When AI can produce outputs quickly and cheaply, the competitive advantage shifts from output velocity to output quality. And quality comes from process—from the thinking, reasoning, and systematic approach that produces those outputs.

The organizations that will win aren't the ones that use AI to produce more outputs faster. They're the ones that use AI to improve their thinking, refine their processes, and build deeper capabilities.

Whether in education or business, what matters is how we learn to learn. In an AI-native world, process is the moat.

Do this for your team!

Start tomorrow. In your next one-on-one, ask "How did you arrive at this?" instead of "How's it going?" In your next project review, focus on the thinking behind the deliverables, not just the deliverables themselves.

The shift from output-focused to process-oriented management isn't just about better evaluation. It's about building teams that can think systematically, learn continuously, and create lasting value in a world where outputs are increasingly commoditized.

I'm very grateful for my undergrad math teachers. Surprisingly, the most valuable lesson they might have taught me is that everything starts with the validity of the reasoning.