§7.1 · Train the Engine, Not Just the Output

Work Capacity Over Window Dressing

There’s a difference between looking strong and being strong. You see it in the gym all the time: the lifter who chases their one-rep max every week, treating PRs like progress reports. Impressive, until the plateaus hit or the pain sets in. What you don’t see as often is the lifter dragging a sled after squats. Doing walking lunges in the cold. Rowing in the morning and hitting their accessories at night. That’s the engine work. That’s capacity.

“If you want to go far, build the lungs before the legs.”

In the training world, this is called GPP, General Physical Preparedness. It’s not just a warmup or conditioning. It’s the foundation for everything else.

GPP improves recovery between sets and sessions. It builds joint integrity, tissue resilience, and work tolerance. It makes your system ready, ready to train harder, recover faster, and adapt more consistently. And yet, in many strength circles, it’s dismissed. You’ll hear things like:

  • “Combo sets are my cardio.”
  • “I don’t have time for fluff. Just the main lifts.”
  • “If it doesn’t add pounds to the bar, it’s a distraction.”

That’s the specialization trap. And it’s seductive.

Specialized programs promise fast results. Just do the thing you want to improve, more of it. Until it stops working — until your joints hurt, your energy fades, your momentum disappears, and you burn out.

Repetition without variation isn’t just physically risky — it’s psychologically draining. Without variation, even the most disciplined lifter starts dreading the next workout. Even the most motivated developer begins to disengage.

We’ve seen this in teams that only work on performance. Or only ship new features. The work becomes mechanical. Morale suffers. Curiosity fades. They’re strong in one motion, but brittle under real-world load.

That’s why smart teams, and smart lifters, build in variation. Like the 6/2 cycle popularized by Basecamp: six weeks of deep, focused work followed by two weeks of cooldown. Capacity training, not a break. We explored this in Chapter 6, and other companies have since adopted and evolved it because it works.

GPP isn’t glamorous. But it’s what lets you train hard, and show up strong, again tomorrow.

That’s what Alex Viada understood. A biomedical engineer and hybrid athlete, Viada rejected the false binary between strength and endurance. He trained for both, and thrived. While most lifters feared that conditioning would eat their gains, Viada built an aerobic base robust enough to support heavy squats and sub-5:00 miles. His results weren’t just impressive. They were repeatable. Durable.

You see the same principle in product teams. Linear didn’t try to dazzle with a full feature suite out of the gate. Instead, they trained their engine: performance. Every click, every transition, every load state was built to feel instant, even offline. That invisible investment became their advantage. Users stayed focused, fast, and frictionless. And while competitors stacked features, Linear scaled loyalty.

But the obsession with output can take an even darker turn, in the gym and in the market.

Some lifters now chase not even performance, but the appearance of performance. Injections like synthol or site-enhancing oils artificially inflate muscle volume — biceps, shoulders, traps — to create the illusion of strength. No increased output. No increase in health. Just ballooned tissue under the skin. These are vanity muscles, pure and simple. And in too many cases, they don’t just harm the lifter’s reputation. They destroy their body. Heart failure. Sepsis. Amputation. Even death.

When looking strong matters more than being strong, the mission is already lost.

And in product? It’s not so different.

Teams inject vanity metrics into their roadmaps: flashy dashboards no one uses. AI features for the pitch deck. Slideware that outpaces software. It might look good in the moment, but behind the scenes, the engine’s breaking down. We will come back to the AI version of this problem in the chapter on AI — when the model can generate any output you ask it for, the question of what’s actually under the hood becomes the only one that matters.

You can’t fake work capacity. You can fake a feature. You can fake a max lift. But you can’t fake the foundation.

Burnout is what happens when you try to specialize your way out of exhaustion. Capacity is what happens when you train for the long haul.

The ones who last don’t just train the output. They train the system behind the output.