A few weeks ago I wrote that writing the code was never the hardest part of building a successful product. The obvious next question is: then what is?

This is my attempt at an answer.

Code is cheap now

Pick any SaaS you use every day. Someone with Claude Code and a free weekend can build a working version of it. Not a perfect copy, but one that works.

I know this because it's how I build. I'm a computer science student, and most of the code in my products wasn't typed by me. I describe, I review, I test, I fix. The building part, the part that used to take a team and a year, now fits inside a monthly subscription.

That changes something important. A moat used to be the thing that made you hard to copy: a team of engineers, months of development, money to pay for both. Today, if the only thing standing between you and your competitors is that you know how to build it, there's nothing standing there at all.

So the moat moved. It moved to the things AI can't do for you. Here are the ones I believe matter most: three I learned the hard way, and a few I'm still learning.

1. Choosing the problem

This is the hardest decision you'll make, and AI makes it more dangerous, not less.

When I built Mawq3i, I built it for a problem I had never lived. I had never run a store. I had never heard a store owner complain about their platform. Everyone I showed it to liked it, and almost no one needed it. It was nice to have, and nice to have doesn't pay.

When building was expensive, you were forced to think hard before you started. Now building is so cheap that it's tempting to build first and ask later. That's how you end up with a beautiful product nobody needs.

My rule now: I only build for problems I've felt myself, or heard people complain about without me asking. Video editing passes that test. I edit for clients every week, and I keep hearing the same sentence from people around me: "I wanted to get into content, but editing is what stopped me."

2. Distribution

Here's the uncomfortable truth. Someone with 10,000 people who trust them can launch an average product and have paying users by the end of the day. Someone with a great product and no audience launches to silence.

When anyone can build, the scarce thing isn't the product. It's attention.

That's honestly a big part of why I'm writing this. I learned that if you start building an audience on launch day, you've already started too late.

3. Unit economics

This one is new, and most people building with AI aren't thinking about it.

In traditional SaaS, your thousandth user costs you almost nothing. In an AI product, every time someone clicks a button, you pay for it. A heavy user can easily cost you more than they pay you.

When I was thinking about pricing for my AI video editor, a lifetime deal came up as an idea. It sounds great for getting early users. Then you do the math: someone who pays once and uses the product forever means you lose money on your most loyal users, every single month.

If you don't know what one user costs you, you don't know whether you have a business.

The rest, briefly

Selling. Admiration isn't a purchase. Every business owner who loved Mawq3i and said "maybe in a few months" taught me that.

Taste. AI gives you something that works. It doesn't give you something people love. The difference is in the small details nobody asked for.

Knowing enough to catch the mistakes. You don't need to write the code, but you need to know when the AI just did something dangerous with your users' data, your security, or your performance. This is where my degree still earns its place.

Learning speed. Not how fast you ship, but how fast you learn from the people using what you shipped.

So, is SaaS still hard?

Yes. Vibe coding lowered the barrier to entry. It didn't lower the barrier to success. If anything, it raised it: more products than ever, competing for the same customers and the same attention.

The opportunity is bigger than it has ever been. But the people who win won't be the ones who can build. Everyone can build now. They'll be the ones who combine building fast with an audience and a real understanding of the people they're building for.

Which of these is your weakest? Reply and tell me. I read every one.

Osama.