If you are a web content editor, you work in a CMS every day. You publish pages and blog posts, embed images and videos, add categories and tags, connect profiles and services, array the site structure and URL hierarchy, and so on.

There can be no doubt that the content management system has made life easier for everyone frequenting the web: The developers build self-service systems, the editors create and publish content, and the visitors consume their chosen content in a predictable fashion.

But how did it all start? When did the world wide web transition from manually updating and uploading HTML files to the complex editorial systems of today?

While you’re at it, find out what headless CMS fits your organization:

Checklist: How to choose the right headless CMS

B.C. (Before CMS)

Yes, there was an era when there was no CMS. You can call it the stone age, also known as the early 1990s. In this period the first websites started to appear, in the form of handmade, static web pages built on simple and flat HTML text files—which in turn were copied using an FTP program to a directory in a running web server.

Technological innovations lead to an increasingly visual and function-rich web. In 1993 the web browser Mosaic started to support images, while the introduction of Server Side Includes (SSI) let you separate portions of your site like header and footer from the main content. In 1996 Internet Explorer became the first to support CSS.

The next step was therefore to build interactive websites by using a combination of static markup language and dynamic scripting with programming languages like Perl and Python.

But  building, uploading, and maintaining entire websites with content in a still largely manual way was untenable in the long run. The time was therefore ripe for the emergence of a system for automating and streamlining the process—it was time for the content management system.

Read more: History of websites »

CMS cometh

The first CMS-like technologies consisted of using server-side scripting to generate content sent from a server to the web browser. Familiar web development programming languages like Hypertext Preprocessor (PHP) and Active Server Pages (ASP) came into play in 1995–1997, with the addition of Java Server Pages (JSP) in 1999. These server-side script engines made it easier to build dynamically generated web pages, and from here on out the scene was set for the advent of the nascent CMS.

A dynamic revolution occurred in 1997 with the introduction of the Document Object Model (DOM), an API for HTML and XML documents letting you identify and control parts of a document. The DOM let users access and manipulate the styles of HTML elements like the body or a division of a page. A few years later the introduction of Asynchronous JavaScript and XML (Ajax) added to the dynamic revolution where developers could request and receive data to update a web page—without reloading the page.

Between 1997 and 2002 many organizations and companies made their own custom, homebrew CMS with the existing technologies. Whether they were manufacturers, agencies, or newspapers did not matter, the world wide web existed and they had to put content out there. The CMSs of the era was characteristically often adapted to the specific needs of the organization behind it.

However, several enterprise CMSs began to appear in the middle to late 1990s as well, recognizing a market for uniform and professionally developed content management systems. Examples of enterprise CMSs from the period include FileNet, StoryBuilder, Interwoven, Documentum, FatWire, FutureTense, and Inso.

In the same period the web hosting service GeoCities grew to its zenith, becoming the third most visited site on the web. Acquired by Yahoo! in 1999, GeoCities was one of the first web-based CMSs allowing users to manage their websites. The resulting legions of personal, hobby based, and other kinds of colorful GeoCities websites is a staple of the late 1990s and early 2000s.

World of WCM

The early to mid 2000s witnessed the increasingly professionalization and specialization of content management systems. During this period the moniker “WCM” started to show up,  standing for “web content management.” Whereas CMS is the broadest term—designating content management also for intranets, archives, and business operations—the web content management system is solely directed toward the web.

No matter what you decide to call it, the WCM/CMS of the early to mid 2000s started to cater to enterprise and business needs in an increasingly professional manner. Simultaneously, the period saw the advent of open source CMSs, like Drupal, WordPress, and Joomla. Most of the WCMs contained both the back-end and front-end technology of a website, and could handle text, images, and other files to store, display, and download.

Be smart: How to build a fast and modern site with Next.js and headless CMS »

With technologies allowing for dynamic content delivery, the world witnessed the rise of the so-called “web 2.0”: a more participatory, user-generated, and social web. As websites moved from static brochure to more interactive experiences, the need for more frequent content updating and management arose. Thus, one of the central roles of the CMS became to provide the capability for different user roles and permissions in delivering content.

Also, technological innovation led to a steady increase in other CMS features, including previewing, URL handling, RSS feeds, responsive designs, visitor comments, tracking systems, permission systems, drag and drop, visual editors, templating, integrations with e-commerce, analytics, CRM, and ERP, and so on.

Here are the top features in CMSs year by year (courtesy of Dries Buytaert, founder of Drupal):





Static content

WYSIWYG authoring

Social media integration

Customer intelligence

Separate content from design

Dynamic content

WYSIWYG page design


Animated GIFs

Publishing workflows

Collaboration tools



User generated content

Rich media integration

Service-enabled / APIs


Modular architecture

Lead generation tools

Multi-site platform governance








Donning the DXP

DXP stands for “digital experience platform” and represents the evolution of content management systems into complete marketing suites from the late 2000s and well into the 2010s. In a world of digital transformation, a DXP aims to deliver a coherent experience for brands across different digital touchpoints.

A DXP is primarily directed toward enterprises, and may include a classical CMS together with analytics, e-commerce, machine learning, personalization tools, A/B testing, SEO, and—maybe most importantly in the context of today’s omnichannel world—APIs for content delivery to different channels.

While many people found the benefits of the DXPs to outweigh the disadvantages, others did not think so. Being monolithic behemoths, DXPs could be slow to operate for content editors, and generally inconvenient to handle for developers. This lead to the next step in the evolution—something lighter and more agile: the headless CMS.

Heralding the headless

Approaching the 2000s, the coupling between the delivery layer and the CMS became tighter. This used to be convenient for both editors and developers: the former could use all the rich features now associated with CMSs, while the latter could code, test, and deploy both editorial and end-user functionality bundled.

This was a bed of roses for a time, but two new trends of the 2010s disrupted this:

  1. The introduction of new channels in the form of mobiles, IoT, wearables, etc.
  2. The innovation of front-end frameworks

So, content editors wanted their content in all channels, while developers wanted to build solutions with their favorite front-end tools. Together, these two trends made the CMS vendors see the need of a headless approach.

A headless CMS severs the tight coupling between the content and the presentation we have seen in traditional CMSs, instead becoming a database with structured content, ready to be delivered to several channels—whether it is the classical desktop, a mobile app, an IoT device, or something else entirely.

While not a new concept, the modern headless CMSs of the late 2010s and early 2020s differ substantially from the old manifestations. The client in the past was most commonly a website, whereas now the idea is to deliver content to wildly different channels and devices.

Learn more: 5 reasons to go headless with Enonic »

Returning to roots

As a result of the headless popularity, WCM/DXP vendors added web based content APIs to meet the competition from pure headless vendors. This, in turn, saw the influx of the so-called hybrid CMS or more precisely next generation headless CMS—where you can keep running digital experiences with both traditional and headless features.

Many first generation headless CMSs were essentially databases with an API, completely removing editor-beloved experiences like visual page editing, previewing, and tree structures. A next-gen headless CMS thus returned to the roots from a certain point of view—by reinstating these familiar editorial features while still offering all the benefits of a headless CMS, like structured content and completely decoupled front-end.

And this is where we are today.


In this context, it is useful to know about the different types of CMSs:

CMS type


Static website builders

Generating files. Few changes, limited resources in terms of server capacity, fast to serve files.


For organizations looking for a suite. One package where every service is bundled, with little need to customize solutions.

First-gen headless

Point solution, app, service needing content.

Next-gen headless

Complex solution, building customer journeys, websites, and services. Delivering content to services headlessly.

This is the point in the evolution where we are today. We now have a broad range of CMS vendors: pure headless CMS, next-gen headless CMS with front-ends like Next.js, site builders like Wix, and other varieties. We face a fragmented future in terms of channels and devices.

To prepare for this omnichannel reality, as well as digital transformation, your organization should choose CMS carefully.

Checklist: How to choose the right headless CMS

First published 3 June 2019, updated 29 September 2022.

Related blog posts

Get some more insights 🤓

Get started with Enonic! 🚀