Launching a WooCommerce storefront for a Thai manufacturer

November 2022. I was a second-year computer science student at Thammasat, and since August 2021 I had also been RUAMSUK PLATING LIMITED PARTNERSHIP at RUAMSUK PLATING LIMITED PARTNERSHIP — my family's trophy factory in Pathum Thani, casting, engraving, and electroplating awards since 2006. On the first of the month, RS Medal went live: the second WordPress catalog site I had shipped for the company's brands that year, after RS Award in June. Both sites could show products. Neither could sell them.

The next build had a harder requirement. The trophy brand's own storefront — the site that became RS TROPHY — had to take orders, which meant a cart, a checkout, and everything that stands behind them. I built it on WordPress and WooCommerce. This post sits at the decision point; the project's recorded milestones run from January to April 2023.

//Why WooCommerce, when I wanted to write code

The honest starting point: I did not want to use WooCommerce. I was a computer science student with a working knowledge of the stack underneath it, and the pull toward building a custom cart was real. I picked the plugin anyway, for reasons that had nothing to do with what I felt like learning.

A checkout is not a feature; it is a liability surface. Payments, pricing, stock, shipping rules, refunds — every one of those is a place where a bug costs the business actual money, and I was one part-time developer working between lectures. WooCommerce carried a decade of other people's handled edge cases. A proven commerce engine operated carefully beat a clever one operated alone.

There was also a fact about the company that outweighed anything about the stack: it had been making and selling awards for sixteen years before I added a buy button. The storefront's job was to fit an existing business, not to showcase what its developer could build. That framing settled most of the arguments I was having with myself.

• One person, part-time. The whole build — hosting, theme, plugin stack, content, marketing wiring — had to fit around a degree.
• Money paths need maturity. Cart and checkout flows came battle-tested instead of hand-rolled.
• The ecosystem was the feature. Marketing integrations, analytics, and ad tracking all had beaten paths into WooCommerce.
• The pattern was proven. Two WordPress catalog sites were already live; the storefront extended a system the company already ran on, rather than introducing a new one.

//Four months, in the order the timeline records

The build ran in a straight line, roughly one phase a month. Discovery and setup meant hosting, a theme, and the commerce plugin stack. Design meant the storefront theme and the catalog browsing experience. Development meant the product catalog, the cart and checkout flows, and the marketing integrations. Then go-live — the storefront that ran until the modern rebuild.

//The launch was mostly not the storefront

A manufacturer's storefront does not get traffic by existing — nobody types a trophy factory's URL from memory. Orders start as searches. So the go-live work that mattered most was the least visible: on-page SEO so the catalog could rank, analytics to see what visitors actually did, and ad tracking to buy the demand that organic rankings had not earned yet. The two catalog sites had already proven that playbook earlier in the year; the storefront pointed it at pages with a buy button.

// functions.php (child theme)
add_action( 'wp_enqueue_scripts', function () {
	// Keep cart-fragment polling off pages that have no cart UI
	if ( ! is_woocommerce() && ! is_cart() && ! is_checkout() ) {
		wp_dequeue_script( 'wc-cart-fragments' );
	}
}, 20 );

That snippet is illustrative, but the principle it encodes was the real skill this launch taught: you do not rewrite a platform like WooCommerce, you operate it — child themes, hooks, and restraint. For someone who wanted to write everything from scratch, restraint was the harder half to learn.

//The case against my own choice

The tradeoffs were real. A plugin stack means someone else's release schedule touching your money path — every WordPress, WooCommerce, and theme update was a small gamble taken on a live store. Performance had a ceiling I could push against but never remove. Every feature was somebody else's opinion first and mine second. And the final verdict is in the project's own timeline: in 2026 I replaced the whole thing with a custom Bun monorepo and consolidated the legacy WordPress sibling sites into it, redirects and all. The critics of the 2022 decision eventually included me.

But that number is also the defense. The storefront ran for three years, and it was the version of the business that earned the rebuild. The custom platform I wanted to build in 2022 did get built — in 2026, by which time there were orders to justify it and a developer who had spent those years learning what a storefront has to do before deciding how one should be written. Three go-lives in ten months — June 2022, November 2022, April 2023 — happened because I kept choosing boring.