How to Spot a Website a Contractor Built in One Prompt

The giveaway signs

Most AI-slop sites share a look. Three-by-two card grids with identical icons. A purple-to-blue gradient hero that could belong to any business on earth. Headlines like “Innovative Solutions for Your Business Needs” that say nothing about what the business actually does.

Look closer and the pattern continues. Stock photos of people laughing at laptops. FAQ sections with templated, generic answers that don’t match the actual service. Meta titles and descriptions that are either missing or copy-pasted across every page.

The functional tells matter more than the visual ones. Contact forms that submit but never arrive anywhere. No real admin panel, just static pages someone has to re-upload to change a price. Mobile layouts that technically don’t break, but clearly were never opened on an actual phone.

Why this happens

A contractor describes the business in one prompt, gets a full page back, and ships it. The tool fills gaps with the most statistically average answer it has, which is why everything ends up looking the same: the same spacing, the same four-feature layout, the same reassuring but empty copy.

What’s missing is the part that takes actual time: checking how the business works, what its customers search for, what happens when a form is submitted, how the site performs on a slow connection. A prompt can produce a page. It can’t produce a decision about your shipping rules or your data model.

What engineering work actually looks like

Real work shows up in the parts a generic prompt can’t fake. For fluxrbot.com we built a neuroblog with 98 articles and 14 separate content tools behind it, each one handling a different stage of research, drafting, and formatting. That’s a pipeline, not a single output.

For vareno.fun, an online store shipping to 27 EU countries, the hard part was never the product pages. It was tax rates, currency display, and shipping logic that differs by country and has to stay correct as rules change.

For veyamobile.tech, supporting 6 languages and a database of around 150 mobile operators meant real data work: sourcing accurate operator details, keeping translations consistent, and making sure the right content shows up for the right country. None of that comes out of a text prompt. It comes from sitting with the actual data and testing it.

A checklist before you pay

A few minutes of checking tells you more than any sales pitch.

• Submit every form on the site yourself and confirm you get a real lead, email, or confirmation back
• Open the site on your own phone, over mobile data, not just a resized browser window
• Ask to see the admin panel or CMS, not only the public pages
• Ask what happens on a slow connection or with a long product name, an edge case a template won’t handle
• Check that meta titles and descriptions differ from page to page and describe your actual business
• Ask what was tested, and in which browsers and devices

If the answers are vague or the contractor seems unfamiliar with their own build, that’s worth more than any visual red flag.

What good collaboration looks like

A contractor doing real work will ask you questions before writing code: who your customers are, what they search for, what happens after someone fills out a form. They’ll show you a draft or a structure before building the whole thing, and they’ll explain tradeoffs instead of just saying yes to everything.

AI tools speed up drafts, research, and boilerplate, and that’s fine. The difference is whether a person reviewed, tested, and adapted the result to your actual business, or just hit publish.

We build sites and tools the second way, with the integrations, data, and edge cases that a single prompt can’t cover.