bangkok is not a cheaper brussels

the profile I never read

Clutch describes my studio as "a small Oostende, Belgium-based SEO company founded in 2021". LinkedIn has it as Studio DLVX, headquarters Warschaustraat 23 in Ostend, skills listed as wordpress, SEO, SEA and social media. I checked both on 4 September 2026. Neither profile says Bangkok. Neither says Next.js. Neither says AI.

I have been working out of Bangkok since 2024, with one partner here, Mameaw. What I sell now is Next.js builds, ecommerce migrations and AI integration. The two pages a buyer is most likely to reach before they reach my own site describe a Belgian SEO shop in a town I left.

I did not notice, because I was busy writing pages. There are 58 live URLs on dlvx.co and every one returns 200. Nine service leaves, six sector pages, twenty case studies, four llms.txt variants, a robots.txt that names thirteen crawlers and lets all of them in. Meanwhile the sentence the market actually reads about me was written by somebody else in 2021 and never touched since.

That is the first thing the move taught me, and it has nothing to do with the move. The description that decides a deal is rarely the one on your own site, and in a city where nobody knows you yet it is the only one in play.

what the competition looks like up close

The Bangkok agency market is neither thin nor cheap. Move Ahead Media runs about 200 URLs with English and Thai mirrors, vertical case studies and a sitemap that updates daily. Primal runs separate landing pages for ChatGPT, for Perplexity and for Gemini. Asia Media runs 75 indexed portfolio pages and 52 insight articles.

Here is the strange part. Across those 75 portfolio pages there is not one outcome number. Asia Media's single public build case lists inputs: 440 products, 27 suppliers, 2 weeks. Its only performance line is a client saying "Wow. That seems so much FASTER!!". The homepage counters, the ones that animate on scroll, render to anything that is not a browser as "0+ projects delivered", "0+ trusted clients", "0+ countries served" and "0% repeat partnerships". Nineteen years of trading, and no individual person is named anywhere on the site.

Relevant Audience puts "20+ years of experience" and "proven track record since 2018" on the same page. One checkable contradiction and every number beside it stops counting.

So the shape of the market is city pages at volume, produced faster than anybody can put proof inside them. None of that means these firms build badly. It means nothing has ever checked them.

Before that sounds like a boast: my own homepage says "since 2017" three times, my Thai landing page says "since 2021" and "est. 2024", and a live search result already quotes back at me "The company was founded since 2021". I was doing the exact thing I had just finished criticising, at smaller scale.

the directories are the market

For "web development Thailand" the top of the result set is not agencies at all. It is Clutch, ensun, GoodFirms, TechBehemoths and The Manifest. For custom software development in Bangkok, TechBehemoths holds the top three slots with three different URL types. For Webflow work it is Clutch plus Webflow's own two marketplace pages, and I am not going to outrank Webflow on Webflow's domain.

I have not run the assistant test that would prove the next sentence, so treat it as inference rather than fact: an assistant asked who builds websites in Bangkok is reading that same shelf, because the aggregators are what is structured and citable.

The trickle shows in my own numbers. In 90 days the site took 269 sessions, and 213 of those are direct, which means me, my own agents and bots. Eight came from Google organic. Four from chatgpt.com. One from clutch.co. One from themanifest.com. Tiny numbers, but the shape is not ambiguous. What sends me a human is a directory or an assistant, not the result page I had been optimising for.

Two facts sit next to that. I have no Google Business Profile. And a local pack renders on 15 of the 15 commercial Bangkok queries I captured. In Brussels I treated a directory listing as a low value backlink, something a junior does on a slow Friday. In Bangkok the listing is the answer, and my listing says Ostend. The move is not another page trying to outrank Clutch. It is to be accurate on Clutch.

publishing a price, and the bill it sends

My web design anchor is USD 9,500. At the live rate that is about THB 312,900. Sierra IT Group puts the Thai professional band at THB 80,000 to THB 300,000, and the commodity tier tops out around THB 44,000. So I sit above the ceiling of the professional band, in a market where the number is almost never printed at all.

Asia Media, Primal, Outsourcify and Seven Peaks publish no price, name no responsible individual and state no repository ownership term on any public page. Putting all three on mine is a real difference, and not one anybody copies in an afternoon, because publishing a price forces you to have one.

It also sends a bill. Once the number is on the page I can never argue it against a Thai reference, because I lose that on arithmetic before anyone reads the pitch. I have to argue it against the European rate the buyer already carries in his head, or against the cost of paying for the same rebuild twice. Which quietly decides who I am allowed to sell to: someone who has commissioned work in Europe, or someone who has already been burned once locally. Bangkok did not make me cheaper. It made me a European price inside a Thai result set, which is a harder sentence to write and a much better filter.

four questions, and the ones I failed

The argument I wanted was "senior led". I cannot have it. Seven Peaks runs "senior-led" as the literal headline on its homepage, backed by 85 percent senior practitioners, 200 people, and Accenture, McKinsey, Bain, BCG, SCB and 7-Eleven logos underneath. One person using that same phrase reads as a small shop borrowing a big firm's line, and a buyer is right to read it that way.

What survives is four questions a buyer can ask any agency in this city and check in four minutes. Does it publish a price. Does it name the person doing the work. Does it hand over the repository. Do its case studies carry a number.

I wrote that wedge down assuming I passed all four. I pass one and a half. The prices are on the page. Handing over the repo is how I work. But the strings "wil", "wilcot" and "delvaux" appear zero times in the rendered text of my homepage, my about page and my Thai landing page. The name exists only in JSON-LD, and in searchVIU's test across five assistants, a price present only in JSON-LD was found by none of them. The copy still says "we" as well, including "we have shipped 60+ platforms", for a studio that is me.

The fourth is worse. Across twenty case study pages there is exactly one quantified outcome, and it is an image caption reading "a 60 percent planner acceptance rate". My best proof asset, a WooCommerce shop migrated to a Next.js 16 storefront in three locales in one week last April and still running, is written up in 456 words with no number in it at all. On the main domain that case study used to 404. It redirects properly now, which took one line and two years of not looking.

the honest state of demand

Last 28 days: 815 impressions, 12 clicks, average position 13.3. Over 90 days, 23 organic clicks, and the only queries with clicks attached are "dlvx" with 8 and "piexels" with 1. Non brand organic clicks are not low. They are zero.

The site does not appear for its own service queries. "web development thailand" drew a single impression at position 40 in 90 days. "webflow development agency bangkok" drew one at position 27. "web design Bangkok" drew nothing at all.

What does pull impressions is other people's names. "unreal gym" brought 188 impressions at average position 7.5. "unreal gym bangkok", 62 at 9.5. "momoamo", 78 at 9.2. That is 328 impressions in a quarter arriving on my clients' brand names, and the two case study pages they land on returned zero clicks between them, because those pages never say what the builds did. My largest non client page ranks at 7.6 on questions like "does claude work in thailand", which is people checking whether they can sign up for a product, not people hiring a studio.

Then the part that is hard to type. The leads table holds eight rows in twelve months. Six are my own QA submissions. Of the two real ones, one is a SaaS vendor pitching me. So the site produced one genuine buyer enquiry in a year, and it arrived framed as a partnership.

the lane I did not plan for

Outbound is the only channel producing anything: 264 emails, 12 replies, 7 opportunities. The split is the interesting part. Bangkok real estate agencies, 134 sent, 4 replies, 1 opportunity. Bangkok design and creative agencies, 78 sent, 4 replies, 2 opportunities. Bangkok production and events, 28 sent, 1 reply, 1 opportunity. And a small lane I almost did not run, 24 emails to Benelux firms with an interest in Asia, which returned 3 replies and 3 opportunities at a 12.5 percent reply rate.

The best performing thing I do in Bangkok is sell Bangkok to Belgians.

That is what I would tell anyone weighing the same move. Relocating does not hand you the local market. It hands you a position between two markets, and that position is worth considerably more than the lower cost base everybody talks about. Which makes the first job on my list not another page. It is one sentence in somebody else's database that still says Ostend.