CMS: from Vignette in 1995 to WordPress, Shopify, and the headless era of 2026

A CMS—Content Management System—is the software that lets you create, edit, organize, and publish content on a website without having to code every page by hand. For someone used to publishing on WordPress, Shopify, or Webflow, the idea sounds so obvious that it seems to have always existed. It hasn't: the CMS as a software category is about thirty years old, with a well-documented industrial history and a series of architecture decisions whose impact is felt for years after you choose.

It's worth placing the concept historically to understand why there are so many such different options, and why the practical question "which CMS do I use?" has a better answer when reframed as "what kind of CMS for what kind of project?"

A three-decade industrial history

Mid-'90s: the era of the expensive enterprise CMS. The first widely recognized commercial CMS was Vignette StoryServer, launched by Vignette Corporation around 1995, originally developed for CNET. It was expensive enterprise software (hundreds of thousands to millions of dollars per license + implementation) sold to large corporations and media companies. Other enterprise CMS platforms of the era: Documentum, Interwoven (more oriented toward digital publishing), Tridion.

That era was dominated by proprietary software, heavy on-premises installation, and prices out of reach for any project that wasn't corporate. A personal web page or a small business built its site with hand-written static HTML.

2001-2003: the open source revolution. Three launches changed the economics of the sector:

Drupal was released as open source software in 2001 by Dries Buytaert, then a student at the University of Leuven (KU Leuven, Belgium). Drupal was initially a message-board system that evolved into a general CMS. Its modularity and extensibility made it a standard for complex sites and government projects (the White House used it for years, as did European ministries).

Movable Type, also from 2001, was for years the dominant blogging engine. It lost significant ground in 2004 when Six Apart changed its pricing model and many users migrated to open source alternatives.

WordPress launched as version 0.7 on May 27, 2003, founded by Matt Mullenweg (then 19, in Houston) and Mike Little (in the United Kingdom). It was a fork of the b2/cafelog project, created by Michel Valdrighi and abandoned after it was absorbed by personal difficulties. Mullenweg and Little took the code, continued it, and built what would within a few years become the most-used blogging platform in the world.

Joomla appeared in 2005 as a fork of Mambo (a 2000 project) following disagreements over governance.

The three—Drupal, WordPress, Joomla—made up the "Big Three" of open source CMS for a decade. The economics of the sector changed radically: free software, accessible installation, a growing ecosystem of plugins/themes/modules.

2003-2010: the era of specialized vertical CMS. Specialized CMS platforms appeared by vertical:

Shopify launched in 2006 (founded by Tobi Lütke in Ottawa, originally to sell a snowboard store). It became the dominant CMS for ecommerce.

Squarespace was founded by Anthony Casalena in his college dorm room at the University of Maryland in 2003. It targeted the market of non-technical users who wanted a nice site without coding.

Tumblr (2007) and Ghost (2013) specialized in pure blogging and digital publishing.

2010s: Webflow and the visual builders. Webflow was founded in 2013 by Vlad Magdalin and Sergie Magdalin. It combined visual design with clean generated code, capturing a niche between developers and designers. Other similar platforms (Wix, which had existed since 2006, evolved significantly; Wix's Editor X; Framer) consolidated the category.

2013-present: the headless era. The concept of a headless CMS—separating content management from presentation, exposing content via an API—emerged as a response to the limitations of monolithic CMS platforms for modern websites, mobile apps, and multi-channel experiences. Milestones:

Contentful, founded in 2013 in Berlin by Sascha Konietzka and Paolo Negri, was the first significant commercial headless CMS.

Strapi, open source, launched in 2015.

Sanity launched in 2017 with a more structured approach to data.

Mathias Biilmann, CEO of Netlify, popularized around 2015-2016 the term Jamstack (JavaScript, APIs, Markup) to describe modern web architectures that combine a pre-rendered frontend, APIs, and a headless CMS.

In the late 2010s and entering the 2020s, the adoption of headless CMS grew significantly, especially in companies with complex sites, sophisticated technical teams, and multi-channel needs (web + mobile + smartwatch + signage + voice, etc.).

2026: the current landscape. Several categories coexist:

• WordPress remains dominant in volume: approximately 43% of all websites in the world according to data from W3Techs, a figure that has remained relatively stable over the past few years.
• Shopify dominates ecommerce, with figures around 10-15% of e-commerce sites globally.
• Webflow has grown in the mid-market and among design agencies.
• Wix and Squarespace maintain strong positions among small businesses and portfolios.
• Headless CMS (Contentful, Strapi, Sanity, Storyblok, Payload, Hygraph) capture the enterprise market and sophisticated technical projects—around 30% of enterprise sites according to recent studies.
• Specialized CMS: Ghost for serious publishing, Notion as a lightweight alternative for simple sites, Framer for design-focused sites.

The three architectural models: traditional, headless, hybrid

The most important operational distinction for making CMS decisions today isn't WordPress vs. Drupal, but the underlying architecture:

Traditional CMS (monolithic)

The backend (where content is managed) and the frontend (where it's shown to the user) are coupled. WordPress in its usual form, traditional Drupal, and Shopify Liquid are examples. The CMS receives the user's request, queries its database, renders HTML, and delivers it.

When it fits: websites of moderate complexity, blogs, corporate sites, standard ecommerce. Non-technical teams that value operational simplicity.

Limitations: the coupling limits multi-channel flexibility (the same data on web, app, voice, IoT), performance depends on the server, scaling is more expensive.

Headless CMS (decoupled)

Just a content-management backend. The frontend is built separately (with any framework: React, Next.js, Vue, Astro, etc.) and consumes the content via an API. Contentful, Strapi, Sanity, and Storyblok are examples.

When it fits: complex sites with multiple channels (web + app + IoT), teams with dedicated frontend technical capacity, high-performance requirements, the need to migrate frontend technology without touching the content.

Limitations: significantly greater technical complexity, requires independent frontend development, higher initial cost, a steep learning curve for non-technical editors.

Hybrid CMS (decoupled but with presets)

Intermediate categories: WordPress can be used in headless mode with its REST API or WPGraphQL, Drupal with JSON:API. Also platforms like Storyblok that offer visual editing in headless.

When it fits: transitioning from traditional without losing familiar tools, the need for some headless advantages without the full complexity.

Limitations: it tends to be a compromise that doesn't fully capture the advantages of either extreme.

How to choose a CMS without getting it wrong

CMS decisions have long-lasting consequences (migration costs from weeks to months, depending on the site's complexity). It's worth asking yourself honestly:

Who is going to publish, and how often? If you have an editorial team publishing several times a week, usability for non-technical editors is a priority. If it's only updated once a month, other considerations weigh more.

What kind of content? Blogs and static pages work on any CMS. Serious ecommerce needs something specialized (Shopify, WooCommerce, Magento, BigCommerce). Heavy multimedia (video, podcast) has specific CMS platforms. Technical documentation works better on structured CMS platforms (GitBook, Docusaurus).

What technical capacity does your team have? Without developers, headless is an enormous risk. With a strong technical team, headless can be a competitive advantage.

What sustained budget? Beyond the initial cost, there's the cost of hosting, maintenance, security, plugins/extensions, and training. WordPress may be "free," but a professional WordPress site costs thousands of euros per year in serious maintenance.

Multi-language, multi-region, multi-brand? Some CMS platforms handle it well out of the box (multilingual Drupal is very refined), others require complex plugins (WordPress with WPML, Polylang).

Critical integrations? CRM, marketing automation, analytics, e-commerce, internal tools. Verify the availability and quality of integrations before committing to a platform.

Time commitment? Today's CMS should still serve you in 5 years. Companies that change CMS every 2 years lose a lot of time on migrations.

Lock-in risk? What would happen if you wanted to migrate? Open source CMS platforms have simpler migration; very specific proprietary CMS platforms can be hard to abandon.

Common mistakes in CMS decisions

Choosing by trend. "Everyone's on headless" isn't a reason for your small brand with a blog to move to headless. The added technical complexity is only justified by actually using the capabilities.

Underestimating the total cost of ownership. WordPress appears to be free. A professional WordPress site with security, performance, premium plugins, and maintenance can cost thousands of euros a year. The real economics include all those elements.

Installing too many plugins. In WordPress especially, each plugin is an attack surface, a possible conflict with other plugins, a possible performance problem. A WordPress site with 40+ plugins is usually in trouble.

Not considering technical SEO. Some CMS platforms (especially Webflow, Shopify) generate clean code by default. Others (some bad WordPress installations) produce heavy, slow, or poorly structured code.

Forgetting permissions and roles. Without well-configured permissions, any editor can break critical pages. Defining roles from the start prevents incidents.

Not maintaining backups. A site without backups is a lottery waiting to happen. Automated backups, in a location separate from the hosting, tested periodically.

Migration without planning. Changing CMS without a migration plan for URLs, 301 redirects, content, images, and SEO usually costs traffic for months.

Underestimating the learning curve. Headless CMS can be technically superior and operationally far more complex for the team. The quality of the CMS depends on how well the people who publish use it.

The reality of WordPress in 2026

Since WordPress remains the most-used CMS, it's worth honestly naming its strengths and weaknesses in 2026:

Strengths:

• A huge ecosystem: themes, plugins, available developers, an active community.
• A gentle learning curve for editors.
• Low initial costs (especially with shared hosting).
• Extensive documentation.
• Proven success stories at every scale.

Weaknesses:

• Security: being dominant attracts attacks. WordPress is one of the most attacked CMS platforms in the world.
• Performance: without serious optimization, WordPress sites can be slow.
• Plugin hell: excessive dependence on plugins creates fragility.
• Gutenberg/Block Editor (introduced in 2018): ongoing controversy, many users prefer classic editors like Classic Editor or page builders (Elementor, Bricks, Oxygen).
• Hard to scale to very complex sites without serious engineering.

The future: WordPress.com (hosted) has grown. WordPress.org (self-hosted) remains dominant in volume. Significant investment has appeared in making WordPress more modern and headless-friendly. Managed-WordPress platforms like WP Engine, Kinsta, and Pressable simplify the operation in exchange for a higher cost.

CMS and creative operations

For a brand or agency that produces content at scale, the CMS is the last mile of creative work: where the produced content is published to the public. If that last mile is slow, fragile, or chaotic, the quality of the previous work is compromised. Conversely, a good CMS can multiply the output of a good creative team.

That's why this decision, although it seems technical, connects with creative operations: the editorial calendar coordinates what gets published when and through which CMS, the approval workflows ideally integrate with the CMS so that approval and publishing are a single flow, and content production generates the material the CMS distributes.

Polimake isn't a CMS—it's a creative operations platform that coordinates work before it reaches the CMS. Studio coordinates planning, Studio production, Media the assets. When a piece is ready, it's published to the corresponding CMS (WordPress, Shopify, Webflow, headless, etc.). The two software categories are complementary, not competitors.

---

If you lead marketing, technology, or product and you've landed here looking for an answer about CMS, the most useful thing you can take from this article is probably the first question to ask before the choice: what architecture does your project need: traditional, headless, or hybrid? That decision filters the universe of options more usefully than comparing individual features. Once the architecture is clear, the reasonable options within each category are few and the choice becomes manageable.

To complement this, hosting covers the infrastructure layer on which the CMS operates, domain covers the web identity layer, and SEO covers the discipline whose performance partly depends on the chosen CMS.

Quick references

• Hosting — the infrastructure on which the CMS operates.
• Domain — the web identity through which the CMS is accessed.
• SEO — the discipline affected by the choice of CMS.
• How much a CMS costs — the real cost of ownership.
• Content production — the flow that fills the CMS.