There's still a place for the all-in-one DXP

In response to this, even the more 'Traditional' DXP vendors typically now support headless as a feature. I don't just mean 'they have an API', I mean they fully support modern decoupled web development, with features such as GraphQL APIs, SDKs dedicated to development in Node.js and other languages, and sample sites using Next.js or Gatsby.js.

Traditional vendors have also increasingly been marketing themselves as 'Composable', as they either acquire powerful third-party products to offer as part of a suite of products, or utilise their new modern APIs to embrace the concept of composability with other platforms.

Traditional CMS/DXP vendors have always been very open to integration, extension, and customisation. The idea of a truly 'monolithic' system, i.e. standalone and not connected to or integrated with anything else, is exceedingly rare.

Fast forward to today, and nearly every CMS or DXP appears to be 'meeting in the middle'. Traditional CMS with headless APIs, Headless CMS supporting traditional visual web page editing, and all claiming to be Composable and SaaS to some degree, and none of them Monolithic.

For example:

• Kontent.ai launched as a Cloud-Native (SaaS) Headless CMS and is at the core of many Composable DXP architectures. But it was also one of the first to start adding traditional web-focused visual editing experiences akin to a Traditional CMS, through its Web Spotlight features.
• Optimizely DXP began as a Traditional CMS, but it also offers a SaaS-like hosting platform and a larger product suite positioned as a Composable DXP, plus dedicated Headless APIs.
• Sitecore XM Cloud (their brand new platform as of 2023, as opposed to their legacy Sitecore Experience Platform) has gone even further, being rebuilt as a Cloud-Native SaaS Headless CMS, positioned as a Composable DXP, while offering a more Traditional web content management experience via their Sitecore Pages product.
• Xperience by Kentico is a Unified DXP, with a broad set of content and marketing features in one product, but also offers a fully vendor-managed SaaS platform, modern GraphQL headless APIs, and the ability to remove, add, customise or create third-party modules in a highly composable manner.

Are any of these terms still relevant?

Of course, terms such as 'headless' and 'composable' are still important, but with the way platforms are evolving (including their product marketing), these definitions alone should not be used as the cornerstone of your DXP platform selection strategy in 2023. 

It's no longer prudent to draw a hard line between 'monolithic' versus 'composable', or even 'headless' versus 'traditional'. We regularly work with headless, traditional, composed and unified platforms, yet none of them are 'monolithic', and all of them typically form part of a composable DXP via integrations (it's exceedingly rare to see a platform truly standing alone as an all-in-one).

So how should we differentiate?

With all these different classes of platforms starting to blend together in terms of features and capability, the biggest platform selection factors I see in the many Explore projects we undertake focus on complexity and flexibility. In most cases, they go hand-in-hand.

On one end of the scale, you have a fully bespoke, best-of-breed, Composable DXP. Incredibly powerful, highly optimised, and (hopefully) perfectly fit for purpose. Extremely flexible, avoiding vendor lock-in, with the ability to add and remove technologies over time without a complete rebuild. But at the same time, it brings a high level of complexity, in terms of development and DevOps/support, but also in terms of licensing and contracts.

On the other end of the scale, you have a single-vendor, Unified DXP. A broad set of features out-of-the-box limits your choice of tools for many marketing functions, and you're dependent on a single vendor's roadmap to provide most of the features you desire. But you get a single vendor, with a single contract and licence fee, a single point of contact for support, one login for your users, a consistent user interface, and drastically simplified infrastructure and DevOps requirements (especially if it's cloud-native).

(In the middle, you have vendors offering their own 'pre-composed' set of products under a Composable label, which could have been built by them, or acquired and brought into the fold. So while they are separate products with their own licence fees, they are expected to be used together, reducing some of the complexity of a fully bespoke setup, but also some of the flexibility of a best-of-breed approach.)

Breaking it down, most platform offerings today could be summarised as being either:

1. Unified DXP - such as Xperience by Kentico or Umbraco
2. Pre-composed DXP - such as Optimizely or Sitecore
3. Composable DXP - no single vendors by definition, but typically including a headless CMS such as Kontent.ai or Contentful combined with other MACH microservices.

Decision factors – do you really need a composable DXP? 

For example, you might have multiple varied channels in a large enterprise ecosystem, with an advanced Customer Data Platform, heavy investment in an existing Commerce Platform, and you are looking for a cloud-native Headless CMS to fill the gap. You have an advanced internal technical capability, team members who are capable of supporting a MACH architecture, and are keen to build on it with a roadmap of future projects. 

Giddy up, Composable DXP is for you!

On the other hand, maybe you're looking for a platform to drive your web channel(s), and maybe share that content with other channels in the future, such as a mobile app. You want to segment your website visitors for personalisation and A/B testing, and drive email marketing campaigns. You have some key systems in place already, such as a CRM, that you need integrated with your CMS. You want your editors to be able to easily edit your site and build marketing landing pages within your brand guidelines and set up online forms. You are budget-conscious, with limited internal dev resources. Do you need a composable DXP, combining a number of products and architecting a solution to meet these goals? What if there was a single product that met these needs with a single licence fee, was quicker and cheaper to get built, integrated smoothly with your CRM, and was easier for your team members to use and maintain? 

You should seriously consider a Unified (or all-in-one) DXP.

Benefits of a Unified DXP

Pros

• Single vendor, single licence, transparent pricing. You pay once for the full feature set.
• If your requirements are fairly typical, you can get almost everything you need in one place.
• The developer experience is extremely good, thanks to a unified codebase, documentation, support, training, and certification.
• The end user/editor experience is also extremely good, thanks to a unified and consistent user experience across all features/modules, documentation, support, and training. Users also only need to log in to a single platform.
• Clear roadmap and excellent coverage of features. The vendor is considering users' end-to-end experience and requirements, not just a particular feature.
• The partner ecosystem for a Unified platform makes it very easy to find an agency to support your entire platform.

Cons

• Typically a very broad but not as deep feature set as you will get from composing your DXP from a suite of specialised products.
• To make the most of fully integrated features and get the best value for money, you will typically limit your team to using the out-of-the-box features (such as email marketing), rather than your favourite third-party service (while it's certainly still possible).

The common denominators

Remember, there are no monoliths in 2023. These approaches (Composable, Pre-composed, and Unified) typically share a common set of advantages:

• They can all drive a website plus additional non-web channels
• They all support cloud architectures and hosting
• They all have headless APIs
• They are all easy to integrate with third-party systems.

There can be all-in-one ⚔️

Traditional DXP, Headless CMS, Composable and Unified platforms have all reached a level of maturity now that they can compete in common ground for the majority of digital website projects. The decisions are now far less commonly driven by factors such as 'headless' versus 'traditional', or 'composable' versus 'monolithic'. Today, digital experience platforms are better differentiated by whether they are a Composable DXP (composed of many specialised products, often including a Headless CMS), or a Unified DXP (primarily a single product).

If you need a Headless CMS, and/or you're ready to take on a Composable DXP architecture, go for it!

But if your requirements aren't extraordinary, and you're looking for increased reliability, reduced complexity, and excellent value for money, you really should be looking at a Unified DXP.