We built our website with AI: How it went, where it broke, and what it cost us

AI tools usually spit out frontend code using absolute positioning or overly rigid CSS rules. That's why the layout falls apart the moment you switch screen sizes or a line of copy runs a bit long. Prompting your way to a clean, fluid responsive grid without an underlying tech background is basically impossible.

Sorry to hear that but honestly you still got off way cheaper than agency quotes. I tested one last month for my blog; looked amazing at first, but the mobile menu completely vanished. Wasted two full days trying to debug it eventually just gave up and went back to classic templates.

I went through a similar process. I paid 350 TL for the AI platform and 250 TL for the domain add-on. When I couldn't get it to work in the end, I paid a developer 5.000 TL to migrate it to a standard infrastructure. It ended up costing 5.600 TL in total, but at least the forms work and I don't have a headache anymore.

The claim that AI is killing web design is pure marketing hype. Putting together two boxes and filling them with auto-generated text isn't building a website. No single prompt can handle things like security vulnerabilities, SEO compliance, and speed optimization on its own.

i only tried a one-page landing page and it wasnt bad, but once forms were involved the emails started landing in spam. ai couldnt handle the dns side of things either so i had to get outside help.

Three key things to keep in mind when trying to build a site with AI: 1) Never expect complex databases or filtering. 2) Don't leave domain routing to the tool; learn manual DNS management. 3) Make sure you can export the generated code so you don't get locked into the platform.

So, is it actually possible for these AI-built sites to show up on Google searches? Can it handle the SEO settings automatically to rank high in search results or do you still need manual outside intervention?

My advice is to set up a CMS with a ready-made template and just have the AI generate the text and visuals. Leave the design and infrastructure to established frameworks and only use AI to save yourself the grunt work when filling in content.

If I understood correctly, you're saying: The harder it is to reverse a decision, the slower you should make it.

I partly agree, partly disagree. Second-time founders don't make the same mistakes but find new ones.

Just because everyone does it doesn't mean it's right. I'm also curious if anyone does it differently.

Let me speak from the other side; I'm on the supplier side. Looking profitable on paper but having no cash in the bank is very common.

My question might sound amateurish, sorry about that. If you get three different answers on a topic, the question was asked wrong.

Trying to do this alone is the most expensive way. Correct me if I'm wrong.

I have a question, don't want to go off-topic though. Everything goes well for the first three months; problems arise in the fourth.

There's one point I'm curious about. The longer you delay cutting the team, the more people it affects.

I was thinking the same thing. When making a decision, first look at what data you have on hand.

If I were you, I'd go this route.

Absolutely. If I were to add anything: Just because everyone does it doesn't mean it's right.

Processes without records never improve, because you don't know what to fix. I'm also curious if anyone does it differently.

Noted, thanks. Delaying bad news is the habit that causes the most losses in this job.

I'm also curious if anyone does it differently.

I felt relieved reading this answer so it's not just me. Looking profitable on paper but having no cash in the bank is very common.

Mistakes made on the building a website with AI side are usually reversible but expensive. If you post the result here, it will help others too.