Paper pallet exporter — full SEO optimization

A B2B manufacturer that makes paper pallets and exports them internationally. Product info and certifications were well-documented, but search engines couldn't read the site structure, so potential buyers struggled to find the site.

SEO for a B2B export site has a different texture than a typical B2C site. Search volume is small, but almost everyone running those searches is a near-decision-maker — a potential buyer — and a single organic visit can turn into a substantial contract. So what matters isn't keyword volume; it's whether the people specifically looking for your product can find your site cleanly.

This paper pallet manufacturer had product information, technical sheets, and certifications nicely laid out on the site. But once we looked at how search engines were actually seeing the site, signals were leaking in several places.

The first thing we found was missing hreflang in the multilingual setup. The site ran in two languages — Korean and English — but there was no signal anywhere telling search engines that the two language versions corresponded to each other. From a search engine's point of view, the same company's same product looked like two unrelated pages, and in the worst case those pages could end up flagged as duplicate content. Someone searching from Korea should land on the Korean page; someone searching from the US or Europe should land on the English page. That mapping simply wasn't there.

Structured data on product pages was also absent. Without Product schema, search engines weren't explicitly recognising these pages as "pages for a product called paper pallet". Even with rich certification, spec, and material info on the page, none of it qualified to show up as rich content in search results.

Per-page metadata was inconsistent too. The Korean homepage leaned into a "company introduction" tone, the English homepage used a different tone, the product category pages used yet another format — tone and keyword mapping were scattered across the site. The sitemap was an older version, and robots.txt had no policy for the multilingual paths.

We started the work with the multilingual SEO layer. We added hreflang on every page in both directions. For example, /ko/products/eco-pallet declared itself as ko-KR and its English counterpart as en; /en/products/eco-pallet did the symmetric declaration back. We set x-default to the English page so that regions without an explicit language match would default to English. Getting search engines to recognise the two language pages as "the same content in two languages" was the starting point of everything that followed.

Next was the metadata and schema cleanup for product categories and individual product pages. We unified titles into a "{product name} | {category} | {company name}" format, and built the meta description so it auto-assembles the product's core specs, dimensions, and certifications. Product JSON-LD included product name, material, dimensions, manufacturer, and certification info; the company information area carried Organization JSON-LD with business info, contact details, social links via sameAs, and the business registration number.

We rebuilt the sitemap to match the multilingual structure. It now includes every public page under /ko and /en, with the corresponding hreflang information reflected in the sitemap itself. The robots.txt got a clear multilingual path policy, and we blocked private areas (/ko/admin, /en/admin and similar) on both sides.

Finally we sorted out OG and Twitter metadata and finished Search Console registration and verification. On a multilingual site, OG tags also need to vary by language so that social shares show the correct-language preview — we handled that as well.

The directly measurable changes after the work were clear. Bidirectional hreflang mapping is applied across all multilingual pages, Product and Organization JSON-LD pass the validation tools, and the sitemap correctly reflects the multilingual structure. B2B export sites are an area where the structure you put in place once tends to compound over a long horizon, so this cleanup laid the foundation for potential buyers to find the site cleanly whether they're searching in Korean or English.