Composability refers to the ability for different components or elements to be combined or connected in various ways to create larger, more complex systems or structures. It’s a concept often used in the context of software development, computer systems, and engineering, but it applies to other fields as well. Content, for example!
In this post, we’ll explore definitions of composability in relation to various fields of digital work. We’ll also offer a helpful analogy to illustrate how composability works (it’s not LEGO, we promise), talk about composable content platforms, and explore how composability could be further enriched by the innovations made possible by artificial intelligence and machine learning.
What is composability?
Research and consulting firm Gartner first brought composability to mainstream attention with a keynote stating that the future of business is composable. Citing examples of businesses struggling to cope with the demands of the COVID-19 pandemic, they made the case that organizations that pivoted to modularity, autonomy, orchestration, and discovery would be better prepared for uncertainty.
But the term also has parlance in several fields. In software development, composability refers to designing software components or modules in a way that they can be easily combined, mixed, and matched to create new applications or functionality. More commonly known as composable architecture, software is typically built with well-defined interfaces, making it easier to integrate with other components without requiring significant modifications. This modularity and flexibility enables developers to reuse existing components, reducing redundant work and promoting efficient code maintenance. Don’t think of it as a conventional tech stack: call it a tower of power.
In the context of computer systems, composability can refer to the ability to pool and allocate resources dynamically to meet changing demands. For example, in cloud computing, the concept of "composable infrastructure" allows the creation of virtualized resources (such as computing power, storage, and network) on demand, enabling efficient resource utilization and scalability. Check out this glossary by Hewlett Packard Enterprise for more information.
Content is also essential to the discussion around composability. Put the two words together, and you have composable content, which is about the ability to break down content into smaller, reusable components that can be combined and arranged in various ways to create different layouts, pages, or even whole websites and apps. A modern approach to content that emphasizes flexibility, reusability, and scalability in this way is the composable content platform. More details on that later in this post.
The best analogy for composability
At this juncture you might be expecting us to use the example of LEGO and building blocks, right? We promised that we wouldn’t. While plastic building blocks have long been a reliable analogy for composability, we’d like to try something different. The example we’re going to run with is building a personal computer.
Many folks are happy to buy their computers pre-built by major technology companies, whether it’s a desktop machine for the office or a laptop for the coffee shop. But some prefer to tinker and build their own PC. It’s fun! And educational! But mostly fun!
What are the essential components of a PC? We can summarize these as the CPU (central processing unit), the mainboard, RAM (random access memory), the GPU (graphics processing unit), PSU (power supply unit), cooling and storage. Oh, yes, and let’s not forget the operating system (OS).
The joy of PC building is the interoperability of these components — but not all components are the same. Based on the desired purpose of the PC, whether it’s for gaming or productivity, performance or efficiency, you can pick and choose your components from a flourishing ecosystem of hardware vendors. You might want to save on costs, for instance, picking a CPU with integrated graphics and installing an open source OS like Ubuntu.
When it comes to its final form, your PC is more or less a unique confection built to your specific purpose. But here’s the other benefit of building your own PC: At any time, you can swap your components for something else. Let’s say you want to start playing games at 4K resolution: in this case, you’d plug in a top-tier GPU capable of pushing more pixels and hook it up to a 4K display.
These are the principles of composability in action. Indeed, no less a luminary than Gabe Newell of Valve is a fan. Accepting a best hardware award for the almighty Steam Deck in 2022, he said: "So, on my PC, I have an AMD CPU, a Nvidia GPU, the PC's from Falcon Northwest. I have a Corsair mouse, I have a Logitech keyboard, I have a Samsung monitor… And it's that interoperability, that compatibility, that openness that really enables products like Steam Deck."
Connecting theory with practice
So how does this translate into a real-world example of composability that a creative professional or developer might appreciate? Let’s connect theory with practice.
Open up a web browser, visit an online store, and take a look at a product description page (PDP). You might be shopping for a new graphics card for your PC, for example.
A page like this actually draws content from at least three different sources:
Editorial content from a CMS
Merchandising information from an ecommerce platform
Customer reviews from a third-party review service
So, here you have one page that’s *composed* from three different content sources.
Let’s flip it around and take the perspective of a hard-working content editor who has a highly trafficked longform asset in their repository, like a regularly updated guide to GPU performance benchmarks. They want to be able to serve up this one asset to audiences scattered across multiple channels — without necessarily multiplying the workload by a factor of X.
How could this one piece of content be presented?
It will be published in full as a dedicated page on your site…
…but it could also appear on a social media channel like LinkedIn with a title, image, a link, and a share button.
And if you’re using a personalization tool, this asset could also be queued up as a “what to read next” recommendation for visitors browsing elsewhere on the site.
From the one content source, therefore, you are optimized for three different outputs. Composability!
Composable content platforms
Still with us? Okay, let’s circle back to composable content platforms. These offer many benefits, like improved content consistency, faster content creation, easier maintenance, and the ability to adapt to changing business needs and trends. They have become increasingly important as organizations seek more agile and flexible approaches to content management in the dynamic digital landscape.
Traditional content management systems (CMSes) often treat content as monolithic entities, where each page or article is a standalone unit. This approach can lead to inefficiencies, especially when managing complex websites with many pages that share common elements.
The Contentful Composable Content Platform, on the other hand, allows content creators to produce and manage smaller, more granular content components. These components could be as simple as individual text blocks, images, or more complex elements like product listings, call-to-action sections, or forms. These components are designed to be independent and can be mixed and matched to build different pages and layouts.
We won’t list all the key features of a composable content platform here, but we’d be remiss if we didn’t touch upon the API-first approach that underpins everything. A focus on providing APIs (Application Programming Interfaces) enables content components to be easily accessed and used by various applications — e.g., microservices — not just the content platform itself.
Composability and machine learning
Today, composability is cool because it allows you to work across channels and handle a variety of channels from a single platform. But guess what? Tomorrow, composability will still be cool because it makes it so easy to inject new machine learning and AI capabilities into your content lifecycle.
Wait, what? Machine learning and composability? Let’s back up a bit. Machine learning is having a profound impact on composability because it complements and enhances the capabilities of composable content in various domains.
Machine learning algorithms, as a subset of artificial intelligence, can learn patterns and make predictions from data, enabling them to assist in content creation, management, and system design in the context of composability.
The PC building analogy is applicable here as well. A major online vendor in the U.S. has launched a Custom PC Builder which touts an “AI expert to help you build.”
Type in a prompt like “I want to build a VR-ready computer under $2,000” and the service presents you with a list of components capable of exactly that. It’s not perfect — a seasoned builder will find nits to pick — but it’s a terrific proof of concept.
We’re still in the dawn of this exciting era, but here are some other ways in which machine learning is expected to further enhance composability and content production:
Content generation: Machine learning algorithms, particularly natural language processing (NLP) models, can be trained to generate high-quality content components like articles, product descriptions, and reviews. These AI-generated components can then be used as building blocks for further content composition.
Content recommendations: Machine learning can analyze user preferences, behavior, and historical interactions to recommend relevant content components that align with users' interests. These personalized content recommendations can enhance user engagement and satisfaction.
Intelligent content assembly: Machine learning has the potential to optimize content assembly by learning from user interactions and feedback. The algorithms can adapt the arrangement of content components based on user preferences, ensuring dynamic and tailored user experiences.
Content quality assessment: Machine learning algorithms can analyze and evaluate the quality of content components, identifying issues such as grammatical errors, readability, or potential plagiarism. This ensures that only high-quality components are used in content composition.
A/B testing optimization: Machine learning can optimize A/B testing processes by efficiently analyzing the performance of different content combinations and providing insights into which combinations yield the best results.
The pairing of machine learning with composable systems could empower content creators and system designers with advanced tools and insights to deliver personalized, relevant, and efficient content and user experiences.
It opens up new possibilities for content management, marketing, and system development, driving innovation and efficiency in various industries. However, responsible and ethical use of machine learning in composability is crucial to ensure transparency, fairness, and user privacy. Read more about the concept of responsible AI here.
In general, composability means you have far more flexibility, scalability, and adaptability in complex systems.
Don’t think of it as a marketing buzzword but more as a long-term strategy for future-proofing your content. By assembling components or elements within a composable architecture, you can easily rearrange and extend functionalities, leading to more efficient, maintainable, and adaptable solutions.
And if you only take away one thing from reading this post, it’s that you should totally consider building your own personal computer from scratch.