Daedong FE & Build-One
Five company websites designed and shipped in four months, as the group's first marketing hire
Details
- Project
- Daedong FE & Build-One corporate and product websites
- Company
- Build-One Company, part of the Daedong FE group (exterior building materials, Korea)
- Role
- Web Development & Digital Marketing Intern, the group's first marketing hire. Brand refresh, UX/UI design in Figma, front-end build in React.
- Worked with
- CEO, engineering director, one Daedong FE engineer
- Tools
- Figma, Adobe Illustrator, React, TypeScript, Tailwind CSS, AutoCAD 3D, AI video tools, Claude and Codex
Overview
- The problem
- The group had one ageing website. Build-One had none, and the product pages read like engineering documents.
- The decision
- I worked with the CEO and engineering director on a shared order: product, installation examples, specifications, then contact.
- What we made
- Five sites went live in four months. I designed two in Figma and built three from the shared components.
At a glanceFive websites live in four months, from a starting point of one.
01 / Why this project existed
Daedong FE makes exterior building materials: stone panels, ceramic panels, brick systems, and the fasteners that hold them onto a building. Build-One is the group company that sells the brick line. Customers are architects, contractors, and procurement teams. They pick a material by reading the specifications and looking at where it has been installed, then they call sales.
When I joined in September 2025, the group had exactly one website, and Build-One had none. The Daedong site dated from 2019 and looked like it: a stock-photo banner, a template layout, a Google Translate widget in the header, and product pages that read like an engineering document. Someone landing on it could not tell what the company sells from the first screen, and none of the four product lines had a page of its own. Build-One, whose whole job is selling brick, had nothing to send a customer to.
There was no designer or marketer in-house before me. The brief from the CEO was to refresh the brand and give each product line a proper website, in one four-month term.
02 / What I decided, and why
Start with the two sites that mattered most.
In the first planning meeting we ranked the five sites. Two came out on top: the Daedong FE corporate site, because it is the front door for the whole group, and the Build-One site, because Build-One exists to sell materials and had no site at all. Those two got full Figma design passes. The other three product lines would come after, built on whatever the first two established.
Decide the page content in the room, with a reference everyone agreed on.
Page structure came out of meetings with the CEO and the engineering director rather than from a spec. We went through the old site page by page and cut what customers did not need. For the look, the CEO pointed to the website of Haema Architects, a Korean architecture office: clean, premium, simple. Having one site everyone could point at mattered more than it sounds. "Premium" means something different to each person in the room until there is a picture on the table.
Every product page ended up in the same order: what the product is, what it looks like installed, the specifications, and a way to contact sales. A customer who has used one of the five sites can use the others without relearning anything.
Build the last three sites straight from the components.
By the time the first two sites were done, the component library (navigation, hero, product cards, specification tables, project gallery, inquiry form) covered everything the remaining sites needed. So Thin Ceramic, Modular Stone, and SE Fastener skipped Figma. I built them directly in the company's React, TypeScript, and Tailwind CSS codebase and worked out each page day to day with the CEO and the engineering director. The CEO knew what customers ask about each product; the engineering director owned the codebase and reviewed my work. Show a page, get feedback, change it the same afternoon. That loop replaced a formal handoff. All three sites took about a month in total.
I used Claude and Codex as coding assistants during this part, mostly for Tailwind layout questions and for turning repeated patterns into components.
Brand refresh and product videos
Keep the logo, but sharpen it.
The CEO's first ask was a brand-new logo, something simple and premium that broke from the old one. I made several rough directions and brought them to a meeting. Seeing them next to the existing mark changed the conversation. The old logo is based on the spring anchor the company holds a patent on, and nobody wanted to give that up. So the decision became: keep the logo, make it sharper and cleaner, and fix what broke on screens. I then wrote a short usage guide for it (clear space, minimum size, backgrounds, what not to do) so the sites, business cards, and social posts would all use it the same way, and designed Daedong FE's business cards on the refreshed mark.
Build-One's logo stayed exactly as it was. The company wanted to keep using the original, so my job there was to make sure it sat well next to the refreshed Daedong FE mark on the new sites.
Settle the colour argument by splitting it.
Colour took longer than the logo. I wanted gold and navy across the group, because that reads as construction and quality. The CEO kept coming back to burgundy. We went a few rounds before landing on a split that followed the products. Build-One sells brick and fasteners, so its two sites got a warm red. Daedong FE and its two product sites got navy, kept clean and minimal. Each side got what it cared about, and the five sites still read as one group.
Give the sites something to show.
The sites and social channels needed product content, so I made four YouTube Shorts from the company's AutoCAD 3D models using AI video tools, and updated the profiles of the existing Facebook, Instagram, and YouTube channels to match the new identity.
Separately, I worked with a Daedong FE engineer on an AI system that automates construction drawings, and I am named as an inventor on the pending patent application.
03 / Results
All five sites went live by December 2025 and are still running.
- 5 websites designed and launched in 4 months, from a starting point of 1
- 2 sites designed in Figma, 3 built directly in code from the shared components, the last 3 in about a month
- 1 component library reused across all five sites
Other work during the term
- 1 logo refreshed (Daedong FE), with a usage guide and business cards
- 4 YouTube Shorts produced from AutoCAD models
- 1 pending patent application, named as an inventor
I do not have traffic or inquiry numbers for the sites. The company did not track them during my term. What I do have: colleagues who had used the old site told me the new ones were a clear step up, and the links below, which anyone can check.
Two things this page does not show. The sites are in Korean, so the screenshots here have English captions. And a director handled the English version of the ceramic site and the SEO, not me.
What I learned
The hardest part of this project was not the design or the code. It was the feedback, which was often vague ("more premium," "sharper"), and turning that into something I could actually change took longer than any single screen. What worked was bringing options instead of questions. Several logo directions instead of "what do you want." A reference site instead of adjectives. A palette split instead of a palette debate.
Building pages myself also changed how I designed them. After the first site I stopped drawing layouts that would be expensive to build and started designing with the components I knew I had.
