Ask how to get interior design clients in Dubai and most of the advice online is about portfolios and referrals. Fair enough. But a villa owner now also asks an assistant who to hire. The assistant can only use what it can read. So on 5 October 2026 I scanned five Dubai villa interior design studio homepages the way a crawler sees them: raw HTML, JavaScript off, no login.
I expected a story about thin pages. One of the five told a different one.
How do interior designers get clients in Dubai?
Mostly through finished work people can see and through recommendations from people who hired them before. Autocomplete shows owners typing “how to get interior design clients in dubai”, “how to get interior design clients online” and “how to get interior design clients reddit” (UAE locale, 5 October 2026). An assistant’s answer is a newer layer on top of that, not a replacement.
I picked five studios from a web search for Dubai villa fit out and design. Here is what the scan returned.
| What was measured | Result across five studios |
|---|---|
| Readable words with JavaScript off | 1,709 to 4,303 for four sites, 4 for one |
| Sites refusing AI crawlers at the server | 1 of 5 |
| Sites disallowing an AI bot in robots.txt | 0 of 5 |
| Sites with FAQ (question and answer) schema | 2 of 5 |
| Sites serving an llms.txt | 3 of 5 |
No studio is named. This is a finding about the category. Naming one firm’s weak result in public would be a poor way to treat people we may work with.
Can a Dubai interior design website block ChatGPT without knowing it?
Yes. One of the five homepages returned a 403 error to three AI crawler user agents (OpenAI’s, Anthropic’s and Perplexity’s) while serving a full page to an ordinary browser. Its robots.txt named none of them. The refusal came from the server or a security layer in front of it. In a browser the same page held 974 readable words. To the crawlers it held 4.
That is the kind of block a design team never sees. The site looks fine on every phone in the studio. Security tools and hosting settings often turn on bot protection by default. The setting does not tell the owner which bots it stopped.
Another thing to say plainly: a 403 on one day is a scan, not a verdict. Some of these tools respond differently by the hour. What the result does show is that the question is worth checking on your own site.
So what do the other four studios hand an assistant?
A good amount. The four readable homepages served 1,709, 2,621, 2,650 and 4,303 words with JavaScript off. That is enough for an assistant to learn what the studio does and where it works. None disallowed an AI bot in robots.txt and one wrote that answer engines are welcome.
Two carried FAQ schema with real question and answer pairs. Three served an llms.txt file. Headings phrased as an actual buyer question were rare, two in total across all five homepages. Most headings were service labels such as “Villa interior design”. A buyer asks “who can renovate a villa in Dubai Hills and handle the permits”. A heading can only match that if the page says it.
Why does the feed matter if the website is readable?
Because a readable website is only a starting point. An assistant looks for recent and specific material. A studio that posts a finished villa every week on Instagram and a project note on LinkedIn produces exactly that: this community, this scope, this month. A studio that last posted in the spring has an empty feed and nothing fresh to quote. Same neglect, two symptoms.
The reverse holds too. A strong feed pointing to a site that arrives at a crawler as 4 words wastes the posting work. It reaches people who scroll and not the buyer who asks.
Does any of this change whether a buyer clicks through?
Fewer buyers visit a result page when an AI summary sits above the list. The independent evidence is thinner than most marketing blogs suggest. The study worth citing is Pew Research Center, published 22 July 2025, covering 68,879 queries from 900 US adults in March 2025. With an AI summary present, users clicked a traditional result on 8% of visits against 15% without one. Only 1% clicked a link inside the summary. Google disputes the methodology. The sample is American.
What holds up despite the dispute is narrow. If a buyer asks an assistant for a studio and yours is not in the answer, you were not in that conversation.
What should a Dubai interior design studio check first?
Open your homepage with JavaScript switched off and read what is left. Then request it with a crawler user agent using curl and see whether you get a page or a 403. Both take under a minute. Check robots.txt as well, though as this scan shows it is not the only place a block can sit.
The free AI check runs a version of this on your own site so you see your own numbers. After that, look at your feed. If the last project post is from months ago, that gap is the first thing to fix.
The honest caveat
A scan of five sites tells you what a crawler received on one day. It does not tell you whether an assistant would name your studio over another, because nobody outside those companies publishes how that choice is made. Anyone who says otherwise is selling something. We can show whether you have given an assistant something to find. Nothing here promises a ranking or a citation.