Ceysu Kimya: From a Corporate Website to a B2B Digital Growth Platform

Ceysu Kimya needed more than a new coat of paint. We turned an established industrial chemical company's website into a clearer, independently operated platform where product discovery, technical knowledge and B2B enquiries belong to the same journey.

Türkçe sürümü oku
CLIENTCeysu Kimya
MARKETTürkiye · Industrial chemicals
DISCIPLINESWeb engineering · Technical SEO · GEO
53Legacy URLs preservedExisting published paths carried into the new architecture
60Pages in the live sitemapPublished HTML routes, November 2026
7Technical articlesTwo rebuilt articles and five new technical guides
42.6%Smaller active-image setCombined bytes across 39 optimized legacy images

These figures describe delivered scope, not a claimed increase in traffic, leads or revenue.

The challenge: make industrial expertise easier to discover

For a buyer researching water-treatment chemicals, the first interaction rarely starts with a company name. It may begin with a question about coagulants, a comparison of anionic and cationic polymers, or a practical issue in boiler-feed water. Those searches need to lead somewhere more useful than an undifferentiated list of chemicals.

Ceysu Kimya's existing website had valuable service information, published articles and years of accessible URLs. But the journey between technical information, product families, company credentials and an enquiry was not yet designed as one system. The assignment was to modernize that experience without throwing away what had already been published.

We audited the entire footprint before redesigning it

We began with a WordPress content and media inventory, SEO baseline, live URL audit, internal-link review and analysis of real search behaviour. The audit distinguished published pages from WordPress drafts and archives. It also exposed gaps in descriptions, heading hierarchy, canonical signalling and connections between relevant pages.

This changed the project brief. Rather than treating the homepage as the entire product, we mapped how an industrial buyer moves from a specific technical need to a relevant product, service and contact route. Existing SEO value became a migration constraint, while future technical content became a first-class part of the design.

Replacing WordPress without resetting the website

The new site uses a controlled Python publishing pipeline, Tailwind CSS and Nginx-served static output. Source content, design code and generated pages are separated, which makes changes easier to review, test and reproduce. The website no longer needs WordPress to serve the public experience.

The 53 published legacy URL paths were carried forward rather than replaced with arbitrary new slugs. We checked original content, metadata, document references and old media compatibility, then introduced new routes alongside them. The resulting published sitemap contains 60 HTML pages. This was preservation with room to grow, not a destructive relaunch.

A corporate experience organized around buyer decisions

The visual direction brings together a restrained navy palette, industrial imagery, more readable type and less ornamental interface noise. Chemical supply, technical service and consultancy are presented as complementary but distinct paths, instead of making visitors decode the company's full offer from a single generic paragraph.

We rebuilt product and service navigation, made extensive chemical lists easier to explore, and clarified the About page's story and team information. Responsive review across desktop and mobile focused on keeping the technical subject legible and the next relevant action easy to find.

Search architecture was built into the platform

SEO was addressed through page-level metadata, heading structure, internal linking, canonical URLs, sitemap consistency and crawlable HTML. We resolved inherited metadata and archive problems while preserving legitimate published routes. This work gives search engines a more coherent account of the relationships between the company, its products, industries and technical services.

The information architecture connects the search journey: product need, technical explanation, application, relevant service and enquiry. That matters more for B2B discovery than scattering the same high-volume keyword across dozens of generic pages.

Turning a blog into a useful engineering reference

We revised two existing articles, developed five new technical guides and established a dedicated discovery area for technical reading. Subjects include coagulant and flocculant selection for mining, polyelectrolytes, cooling-water corrosion and scaling, and boiler-feed water analysis. These topics reflect real procurement and process questions rather than a schedule of generic promotional posts.

The editorial approach uses comparisons, selection criteria, source references, FAQs and links to the right service pages. A further human editorial review reduced formulaic writing while preserving technical tables, references and URLs. Product-specific specifications, dosing advice and compliance statements remain subjects for qualified technical verification, not invented sales copy.

Making the company more intelligible to AI-assisted search

Alongside conventional search controls, we created machine-readable content counterparts and relevant structured data, with 40 Markdown pages and consistent public discovery resources in the release snapshot. We strengthened relationships between the business entity, products, applications and technical guidance.

GEO here means building a clear, attributable information foundation for AI-assisted discovery. It does not mean promising that ChatGPT, Gemini or Google will recommend a supplier on demand. The platform gives such systems better material to understand, while actual visibility remains a measurable outcome for later reporting.

Giving every image both a job and a budget

Visual work went beyond replacing blurry assets. We introduced descriptive filenames, context-specific alternative text and optimized formats while retaining legacy file paths for compatibility. Relevant content images received structured ImageObject descriptions without fabricating ownership, licensing or certification claims.

Across 39 active legacy images, combined file size fell from about 9.06 MB to 5.20 MB, a 42.6% reduction. The release also includes 68 ImageObject entries. These are measured asset-level improvements, not a made-up site-wide Core Web Vitals score.

The release was engineered, not merely uploaded

The migration used repeatable builds, link and asset sweeps, responsive browser tests, source-content parity checks, sitemap and canonical checks, and isolated release validation. The production workflow kept the Ceysu Kimya website separate from unrelated VPS services and retained a documented rollback path.

We also treated contact as part of the commercial experience rather than a decorative form. The business now has clearer enquiry routes and an operational email-delivery path. The important result is a maintainable customer-facing system that can support further SEO work and commercial measurement.

The outcome: a digital system ready for the next stage

Ceysu Kimya has a modern corporate website that preserves its original published footprint and connects its services to technical content, product discovery and customer enquiries. Its pages are easier to maintain, its content is more purposefully structured, and its publishing process is verifiable.

The project illustrates Mihenk's approach: treat design, software, technical SEO, evidence-led writing and long-term growth as parts of the same engineering problem. Search impressions, qualified enquiries and revenue will be measured over time. The completed result here is the foundation capable of supporting them.

Evidence and scope

  • 53 legacy URLs were mapped and retained during the static migration; 60 published HTML sitemap routes were recorded in the 16–17 November 2026 releases.
  • Seven technical articles consist of two revised and five new pieces; the 17 November editorial release passed its documented browser, source-parity and HTTPS checks.
  • The 39-image reduction of 42.6% measures combined asset bytes, not total website load time, Core Web Vitals or search ranking improvements.
  • 40 Markdown pages and 68 ImageObject records are implementation counts. GEO readiness is not evidence of a measured AI citation rate.
  • No increase in qualified leads, search position, sales or revenue is attributed without a comparable post-release measurement period.

Digital infrastructure built to keep growing.

Explore a project that connects software, search and technical content.

Discuss a similar project