One of the challenges facing customers and analysts (and vendors!) is the ever-changing definition of what a Web Content Management system even is.
Tony Byrne has discussed this fragmentation in some detail. In particular, I like his characterization of “products” (which often do one thing well) vs. “platforms” (which often don’t do a single task as well, but can be customized/configured to do anything). While he does a great job categorizing product/platform against market fit, I’m going to take a bit more of use-case/philosophy angle.
Complicating this fragmentation is the fact that many players who are firmly in one category don’t really want to be in only one category. Enterprise WCM vendors have been adding headless functionality on top to be classified as “hybrid headless”. Web Publishing Platforms such as WordPress VIP or ARC Publishing are now firmly trying to compete with Enterprise WCM in both functionality and client base. Headless vendors also recognize that they need to evolve beyond their well-executed niche of being a great content repository to address marketer needs (while at the same time keeping that simplicity and ubiquity).
David Raab of the CDP Institute did a great service by starting to break down Customer Data Platforms into areas of functionality and running through a few iterations, and I’d like to attempt the same for WCM. This means being somewhat reductionist and drawing some lines in the sand (which may annoy vendors who want to straddle or defy categorization) and this categorization may not be complete, so I’m definitely open to feedback.
Headless / Content-as-a-Service
I’m going to start with the WCM category that some would argue isn’t even “web”. I won’t revisit the definition, but I will provide some commentary to say the rise of this category owes a great deal of philosophy to what became known as the COPE publishing strategy (Create Once, Publish Everywhere – made famous by Daniel Jacobson of NPR in 2009) and it what I consider to be one of the seminal works of thinking in content management.
I do know that Contentful in particular rose out of an in-house need to build a pure content repository for mobile applications, rather than executing a COPE strategy, per se, but the foundations and principles in COPE are perfectly suited for a Headless / CaaS system.
COPE is really a combination of several other closely related sub-philosophies, including: Build content management systems (CMS), not web publishing tools (WPT). Separate content from display. Ensure content modularity. Ensure content portability.Daniel Jacobsen, National Public Radio
The main “big idea” was that you should be managing “core content” separate from display, and this was a big change (and still is – “page management” vs. “content management” is still an ongoing debate in WCM products). And while many other systems could execute a COPE strategy, headless systems are the COPE strategy in the purest form, making the core tasks such as content modelling and creation as easy as possible, without any burdening the user or system with additional features or use cases (yet).
The two typical ways of working with a headless system are either via the API directly (or via CDN) or utilize static-site generators.
Philosophy: Enabling multi-channel content through simplicity and performance, both in architecture/APIs and editing experience.
Static site generators
Strictly speaking, not a WCM system, but I would be remiss if I did not include these here, since they go hand-in-hand with headless systems as a well-established pattern of being able to convert headless content to static output. These include tools such as Gatsby, Jekyll and Next.js
It’s funny, because I was having a chat with an analyst who rightly asked: “How is this any different from the decoupled systems of years ago?” – which is a valid point. I would argue that in the same way that Newtons/PalmPilots were ideas ahead of the technology, the decoupled systems of the past were insanely expensive, architecturally heavy and prone to lock-in (content record storage and transformation to HTML were generally very proprietary and unique to that system – often with custom rendering languages) – whereas the current set of headless and static site generation offers a lot of content portability and framework/programming language selection at a fraction of the cost.
Philosophy: Provide a consistent”programming framework” for getting structured content out of headless CMS and into HTML
Self-service / Customer Portals
Many organizations have web requirements that revolve around unifying disparate content or processes in a single location – be it for employee, or customer self-service. This trend is accelerating especially as systems like cognitive search (e.g. Coveo and others), voice, chatbots, and ML can surface that underlying content from ERP, CRM and knowledge management systems in a faster and more intuitive way.
Like this shift from “decoupled legacy WCM” to headless, portal technology has also undergone a bit of a rebirth from old enterprise-heavy technology standards (remember the Java Portlet Specification?) to something which is a bit more nimble – while still feeling fairly familiar. This category would include vendors such as Liferay and the Salesforce Community Cloud.
Forrester put out a useful document in 2016: “Vendor Landscape: Digital Experience Portals” which does a great job outlining many of the portal scenarios (I am not sure if the 2017 update has this graphic).
Philosophy: Enable self-service on top of underlying knowledge/process systems.
Enterprise Web Content Platforms
Even for someone who spent years in the space, it’s often hard to concisely define something which by definition is often the default. An Enterprise Web Content Platform will include many of the aspects of managing all aspects of content creation for enterprise scenarios (either natively, by add-on module, or via customization). It will often include multiple enterprise subsystems for managing content repositories, library services (check-in, versions, etc.), presentation components and assembly, campaign management, URL routing and management, customer profiles, search indexing.
The “platform” element is key, as this almost requires code to build front-end sites, components and applications (even out-of-the-box components, if included, are usually insufficient to meet all enterprise customer needs). Also, platform implies you can customize “back-end” things like workflow or how the WCM system handles integration with other systems such as DAM.
Typically there is a high learning curve because there is often a high degree of abstraction and technical architecture which is needed to support multiple complex use cases and scalability. In contrast, many other category types would not enable back-end customization at all or within the system.
Philosophy: Architected to enable full customization (of both back-end and front-end/presentation requirements) and scaling of key services such as search.
Web Publishing Platforms
This broad category is aligned with the act of enabling writers to create content and ranges from simple blogs all the way to traditional media publishers emulating forms such as magazine and newspaper articles. Many of the tools and processes are aligned to publisher needs including making writing and linking to related content easy, paywalls, publishing statistics.
I wrote three somewhat related posts which cover many of the Web Publishing platform vendors:
- Part 1: “Some thoughts on Salesforce and Automattic” where I look at the Salesforce investment in Automattic and look at this in the context of those vendors and the market.
- Part 2: “Missing parts of the experience of WP VIP as a full WCM” where I fully articulate the difference between Enterprise WCM and Web Publishing platforms.
- Part 3: “WCM systems are the hot new billionaire status symbol of 2019” where I discuss the publishing/enterprise divide further and ARC Publishing in particular.
Philosophy: Enable effective publishing of long-form content (blogs, articles, etc.)
A value-added host such as Pantheon or WP Engine will provide some advanced hosting capabilities which will take an open-source platform such as WordPress or Drupal and include some enterprise-level features such as more robust hosting and scaling, devops (the ability to automate code builds and deployment) and bundling of common requirements and services.
I hesitated including this category – and I’m sure analysts also struggle with this definition, but it does exist on the fringes of WCM even to the point of including WP Engine (which is built on top of WordPress) being included in many evaluations around Web Content Management. The need definitely is there, but I struggle with separating out the service from the product, but to be fair, even the open-source product/services vendors themselves often have to sort (or actively fight) this out in the market.
Philosophy: Take a well-regarded open-source platform and try to make it more SaaS and enterprise-friendly.
SaaS Site builders
A Site builder is often a good choice for a simple business that wants a web presence with minimal difficulty. It will often handle tasks such as the visual design, domain registration and hosting as well, but usually at an expense of customization and flexibility.
However, this category should not be ignored or dismissed – although they all grew up in the small and mid-sized business (SMB) space, they are all adding more complex frameworks to enable a fair bit of custom development on a SaaS model or better support integration cases (such as with eCommerce). This includes things like adding development platforms such as Wix Corvid or features such as SQL databases and local development.
Philosophy: End-to-end websites for customers with a minimum of code or managing other dependencies
Component Content Management Systems (CCMS)
Originally designed for complex print, documentation and knowledge management scenarios – but having this functionality in a silo is making less and less sense as content itself becomes more multi-channel and relevant to Customer Experience everywhere. If they are not careful, headless/CaaS vendors will eat this space overnight.
Philosophy: Enable technical writers (documentation, knowledge bases, etc.)
It’s not really a category, per se, but there are some WCM/DXP vendors who have found a solid market niche providing experience management on top of the underlying services of other vendors (typically commerce). These vendors (which would include the likes of Bloomreach , Amplience, and Magnolia) provide typical WCM services, but also accept that “best-of-breed” is very much a growing trend and instead focus on interoperability and differentiators (such as search merchandising in the case of Bloomreach) rather than building category-competing features into the product.
(It may be that the only differentiation here between this and DXP is that DXP vendors will often have more bundled into the suite such as Commerce and analytics at the potential expense of interoperability)
Philosophy: Unify and enable other best-of-breed technologies to orchestrate customer experience
Digital Experience Platforms
DXP is still largely considered by many to be a “superset” of Enterprise Web Content platforms (taking WCM as a base and adding elements such as CDP, Personalization/Optimization and marketing automation – for example, Sitecore has Experience Manager and Experience Platform with roughly that breakdown). But this definition of “WCM+goodies” is becoming less true as companies and analysts realize that many other channels are more central to their operation than WCM.
For example, both Gartner and Forrester rate Salesforce as a Leader in their DXP evaluations, yet their WCM capabilities are very new and still quite limited. Many retail stores have moved to a “clicks and mortar” model where the Commerce engine and omni-channel experience is central to CX rather than a WCM – so “digital experience” can really come from a number of different spaces depending on your priority (which is why companies such as Salesforce and SAP now figure highly in these reports which were previously solely dominated by WCM vendors).
Philosophy: Be the core system for everything related to digital customer experience (whatever that may be for your vertical).
“Decoupled” / Legacy platforms
Most of these came about in the age of needing to have a system for managing content at scale within a large organization and have complex content storage and workflow capabilities to match. This would include the likes of OpenText TeamSite, Oracle WebCenter (formerly FatWire) and others. Many have added “live” capabilities, but in many ways they are saddled with a lot of infrastructure and patterns that are overly complex for current marketing operations. For example, in headless systems, content modelling is a drag-and-drop operation in the browser. In TeamSite, it was (is currently?) a very large XML file made by hand in a complex system.
Philosophy: Some organizations invested heavily in these platforms, training hundreds (thousands?) of users and moving off of them would be too difficult. In the same way that Boeing built the 737 Max as a mash-up of old frame with new engines to try to address new competition while avoiding retraining the end user, there is still good money to be made in keeping these accounts supported. (Metaphor provided without comment).
To be clear, many of these categories and vendors are overlapping – and I may even be missing a few (for example, is there a “generic” category beneath Enterprise Web Platforms which is more “product-like”? Mid-market products? Who would the vendors be in this space? Does it even have a “philosophy?”) but I wanted to at least try to get to the heart of some differentiation of the product philosophy behind a few of the major players to help customers in understanding the landscape.