What you choose to build now matters more than how well you build it. (and it always did)

You can build almost anything in a weekend now. An AI-assisted solo founder can ship what used to take a team months. The execution gap, the one that separated "I have an idea" from "here's a working product," is closing fast.

And yet. Most of what gets shipped disappears without a trace.

The reason isn't usually poor execution. The execution is often solid, sometimes impressive. The reason is that the idea underneath wasn't strong enough to carry the product once it existed. Nobody needed it enough. Or fifty others built the same thing the same week because the idea was that obvious.

In photography, beginners fall into a predictable trap. They obsess over gear. Sharper lens, better sensor, more megapixels. They assume better equipment produces better photos. It produces sharper photos. Sharp and good are different things.

What makes a photograph compelling is what the photographer pointed the camera at. Where to look. When to shoot. What to leave out. The camera is execution. The eye is the idea.

AI just gave every builder the best camera in the world. The differentiator moved entirely to what you point it at.

The Catechism That Expired

"Ideas are worthless. Execution is everything." For decades, this was the startup catechism.

I never bought it.

A bad idea executed brilliantly always produces worse results than a good idea executed adequately. The execution crowd confused "necessary" with "sufficient." Yes, you need to build. But what you choose to build sets the ceiling. Flawless execution on the wrong idea just gets you to a dead end faster.

The pivot crowd will point to Slack, which started as a failed game, or Instagram, which began as a cluttered app called Burbn. Both teams eventually found something great.

But the original ideas still failed. What those teams discovered were better ideas, and the better ideas won. Execution can help you find a good idea, but it's a costly way to get there, not a strategy to rely on from the start.

This was always true. But when building was expensive and slow, execution looked like the bottleneck. It consumed so much time and money that people mistook the constraint for the game itself. AI stripped that illusion away. When building takes an afternoon, the quality of the idea has nowhere to hide.

What Makes an Idea High-Leverage

When people hear "ideas are the new leverage," they picture a eureka moment. A flash in the shower. Lightning in a bottle.

That's the wrong image.

Some high-leverage ideas solve a pain people already feel. They have a built-in audience. They're specific enough to be opinionated. And they can't be replicated by anyone who types "give me startup ideas" into a chat window. Good execution still matters here. It's what multiplies the idea's potential. But the idea sets the range of what execution can achieve.

But some of the most powerful ideas don't start with an existing market at all. Nobody was asking for an iPad before Apple built one. Tablets had been tried before, and they'd failed. Apple's insight wasn't "build a tablet." It was a bet on what a tablet should be: touch-first and content-focused, no stylus, no desktop OS. Microsoft had the engineering to build a tablet. Apple had the sharper idea of what a tablet should feel like.

Both types share one thing. They emerge from direct contact with reality. The first kind from friction with an existing problem. The second from noticing patterns in how people actually behave, and making a bet that a new category wants to exist.

AI generates ideas too. Ninety per minute. But AI ideas come from the outside looking in. They pattern-match on what already exists. They lack the texture that comes from living inside a problem or sensing a shift before it has a name.

The Real Time Allocation

Here's where the culture got it wrong, even before AI.

The popular image of a great founder was someone shipping fast, head down, grinding through code and customer calls. The celebrated metric was speed.

But good entrepreneurs always spent serious time thinking. Choosing what to build, who to build it for, which problem to solve, which version of the solution to pursue. The culture just didn't celebrate that part.

Even the best founders spent most of their time building, because execution consumed so much time and money that it dominated the calendar. But the fraction they spent choosing what to build, even if it was a small share of the hours, carried disproportionate weight. That thinking determined whether the building produced anything lasting.

AI made this imbalance impossible to ignore. When building takes hours instead of months, spending a week on what to build doesn't look like procrastination. It looks like the highest-leverage activity available.

The ratio should probably flip. More time thinking, less time building. Because AI compressed the building so dramatically that the thinking phase is where almost all the differentiation lives.

What Investing in Ideas Looks Like

It's a discipline.

  1. Live inside the problem. Use the broken tools. Do the manual workflow. The frustration is the data. Reading about market gaps from a distance produces ideas that sound right and feel hollow.
  2. Kill ideas on paper before building them. Write the idea in one paragraph. Who is this for? What pain does it address? Why hasn't someone solved it already? If the answers aren't clear in writing, the idea isn't ready. Building won't fix that.
  3. Look for signal before writing code. This is harder than it sounds. A landing page with no traffic proves nothing. A tweet can get engagement from people who'll never pay. Conversations mislead when you ask leading questions. Every validation method has a way of telling you what you want to hear. The discipline is in designing tests that can actually say no, and then listening when they do.
  4. Stay with the problem longer than feels comfortable. The first idea is almost never the best one. It's the obvious one. The better idea usually hides behind it, visible only after the obvious one has been examined and found wanting.
  5. Make the risky bet explicit. If the idea requires a market that doesn't exist yet, name that bet out loud. "I believe people will want X because I'm seeing Y." Risky insights produce some of the highest-leverage ideas, but only when the risk is conscious.

None of these look like what "hustle" culture celebrates. They look quiet. They look like someone sitting at a desk, writing, thinking, deleting.

Faster Failure Is Still Failure

There's a trap hiding in the new speed.

AI lets you build fast, so the temptation is to "just try it and see." Ship a prototype, test it, move on. Sounds lean. Feels responsible.

But there's a difference between testing a well-formed hypothesis and throwing spaghetti at the wall at high speed.

When execution was slow, each build cycle forced a pause. The slowness was a crude filter. You had to believe in the idea enough to spend months on it. That screened out some of the worst ideas naturally.

AI removed that filter. Now you can ship a half-baked concept in a weekend, get zero traction, shrug, and start the next one Monday. Repeat fifty times. Each cycle feels productive. None converge on anything real.

Speed without idea quality is just faster failure.

The Compound Effect

Good ideas compound. This is the part nobody discusses.

A well-chosen problem produces more than one product. It produces a direction. It attracts an audience that cares about the same things you do. It generates follow-on ideas that build on each other. It creates coherence across everything you ship.

A mediocre idea, perfectly executed with AI, stays isolated. Each new product starts from scratch because nothing connects to anything bigger.

The leverage gap widens over time. A good idea doesn't just edge out a mediocre one in month one, then plateau. It pulls further ahead, because compounding needs a foundation worth building on.

The camera is in everyone's hands now. The question is what's worth pointing it at.

If this resonated, you might also enjoy: