What Lies in the Future of Content Management Systems?
It’s not just web anymore. Companies need to publish their content to a much larger number of channels and provide the digital practitioner with easy-to-use tools to engage with customers. But how can you assemble services that are highly reusable and highly relevant?
Note: This article was written as a summary of Mark Grannan’s presentation called It’s the End of Web CMS (And I Feel Fine), which was the closing keynote at Kentico Connection Brno. Mark Grannan, Senior Analyst at Forrester, helps clients understand digital experience solutions to drive better engagement with their customers and partners.
What We Can Learn from History
In the late 1800s, Thomas Alva Edison waged a "War of Currents" over whose electrical system would power the world—Tesla’s alternating current or Edison’s direct current system. It was a corporate war for standards, and it took roughly a century to illuminate the world using electricity. Nevertheless, it’s now possible to light up the globe with only one application. To give an example, we can mention one popular fitness tracking app that was released in 2009, Strava, and see that adoption can happen incredibly fast if there are standards in place:
The way we might quantify that is through the notion of value. Have you ever heard of Metcalfe’s law? It states that "the value of a network is proportional to the square of the number of users of the system." So as the number of people using a service increases, the service becomes more valuable to the community. Or we could simply say that when we’re isolated, we grow linearly. When we’re connected, we grow exponentially. When we’re together, connected, the value can radically outstrip the cost after we cross a certain threshold.
How to Get More Value Out of Content
For all those who are building applications and creating content, the value versus the cost is still on the linear curve; it’s under-performing its potential. The challenge here is to rethink how much content and how many experiences we can create on the network. Let’s say that you’re a company with both a B2B and B2C go-to-market strategy, and you have four stacks—B2C Marketing site, B2C Authenticated web app, B2B Marketing site, and B2B Authenticated web app. The website and the authenticated web app have very little in common and don’t share a common backend. It would be good to unify some of those, right?
If you place all the marketing sites on a common stack and the web applications on a different one, you’ll get two platforms, four applications. By bringing all the applications onto a common stack, we get one platform with four applications, which is starting to be a powerful return on investment:
In a modern scenario, it’s not just web anymore. There are a lot of brands and even industries that are simply not omnichannel—they may have a little presence on social media, but it’s not really connected to the backend for web. It would be great to publish out to any number of touchpoints in a really fast, iterative fashion. So, if we unify the two segments and accept for a few channels that aren’t relevant to the B2B and B2C line, we get one platform with a phenomenal value:
If we go one step further, we can have ten segments and ten channels and run all of that from a common backend. When we take into account the time and space dimension, it’s no longer a ten-by-ten matrix, but rather a 3D Rubik’s cube. This level of complexity on a single, vertically-oriented stack gives us the blue screen of death. The system simply crashes, our applications and operations cannot scale to that level of flexibility. So we have to reboot. By looking at a horizontal set of services, we can compose any experience and scale to any number of visitors.
The Reference Architecture Relevant to Any Business
Firstly, we have to start with the customer and focus on the touchpoints. (Starting with the internal operations would be doing it backwards.) We also need the services to build our app. When we bring in the infrastructure—the development stack—and bookend it with analytics and automation, we’re starting to get the shape of any modern application.
We also need to have common core services. This is where our content lives—it is a core service within our application that can be served to any channel and is of similar importance to customer data, the unicorn we’ve been chasing to understand the customer. And we shouldn’t forget transaction services because that’s where we deliver the value—we don’t just have to rely on our e-commerce site and can sell our products through social or a third-party retailer.
The piece that we haven’t mentioned yet and can see in the picture above is the orchestration. We can all curate those back-end services, but they mean nothing if the practitioner (e.g., the marketer, customer service agent, or third-party sales rep) is not enabled to. The practitioner needs contextual tools because they won’t know which segment to pull from or which content’s relevant, so they have to be provided with the right applications. But how to assemble a horizontal set of services that are highly reusable and highly relevant to the practitioner?
With this approach, this reference architecture, you get to be relevant to every aspect of the organization. However, it relies on having a notion of duality: you are the core platform services (content, customer data, and transactions) that maximize scale and quality across channels and systems, and you’re also the front end, the channel-specific applications that allow the practitioners to engage with customers where, when, and how they prefer.
What the Future Holds
The future lies on a headless foundation, although headless may sound scary at the beginning. You might lose WYSIWIG, web publishing workflows (the waterfall model), "in situ" marketing and merchandising rules management, targeting, and analytics. How do you create contextual relevance when you can’t right-click and add a segment, or create a variant directly through the interface?
Although this approach seems quite terrifying and abstract, the opportunities might just be worth it. The ultimate goal is effective and smart use and reuse of content across any channel. But the backend has to be consistent, and that’s why headless is so important.
Not everybody can move to headless right now, but there is an intermediary step that many are taking: cross-channel, splitting apart the legacy "create and deliver" model and moving to a repository that is agnostic (e.g., bringing in a new set of APIs and delivering out to a mobile app from the core system). We can even start to funnel in user-generated content if we have ratings and reviews on social, and we can deliver that back through a campaign engine out to a social experience. That’s possible with today’s vertically oriented stack—the APIs and extensibility have gotten good enough to start aggregating our content.
The Time of Agile CMS Is Coming
To get to the next phase where we cross that threshold, value to cost, we need a new strategy in which we have aggregation, planning, collaboration, and insights in one place. It’s going to allow the creative team, the development team, the channel team, and others to consistently interface with content, which builds on headless in a very literal way. There’s now a notion of an agile CMS, which consists of four basic parts:
- Content hub: Agile systems will unlock the strategic value of existing content across repositories.
- Project management: Built-in planning and curation will accelerate time-to-market.
- Content services: Standardized creation, delivery, and analytics will accelerate optimization cycles.
- Platform: Modern architecture will directly or passively enable Agile development.
However, this is not a solution you buy and immediately get value on. You absolutely need to embrace organizational change. If you can’t get your teams to accept a new model of reporting to each other, planning and collaborating together, then you won’t get the full value of the content. Agile change management, like creating smaller teams with smaller chunks of content, is the greatest hurdle. To illustrate the point, let’s look at Spotify. The parts of the user experience are not assembled by a component within the app; they’re all assigned by teams that are responsible for different parts of the UI. This is why Spotify scales to so many channels with such a dynamic backend—the content is literally changing every minute. To get the value, you have to industrialize your process, move to something that can scale.
One way to do that is through preparing your content for omnichannel customer experiences and achieving a deep understanding of it. We can also embrace "content intelligence", which is the use of artificial intelligence technologies to understand and capture the qualities inherent in any content—its emotional appeal, subject matter, style, tone, or sentiment. There’s a cognitive load on the practitioners (e.g., agency professional, creatives, customer service representatives) that need to make decisions on how to tag content individually, day to day. Artificial intelligence alleviates some of that burden as it allows you to get to a new level of complexity and volume.
Specifically, if we’re trying to sell a new pair of shoes to somebody named Bart, we probably know his profile details, maybe his transaction history, previous purchases, and behavioral data, such as browsing, search terms, dwell time, etc. What do we know about the shoes? We have some product data (like color and materials) and marketing data (the season, use, promo images). How do we cross-reference those to make sure we’re recommending the right shoes to Bart? How are we supposed to link these when we don’t have enough common data or taxonomy? (Or maybe we don’t have the contextually relevant variant of that piece of content.)
Many consider personalization to be a strategic priority for their brand or company, so we shouldn’t let content be the personalization bottleneck. Using variants, dynamic composition, analytics, and attribution, and with smarter content creation, we can optimize the content and learn from it. Imagine it as a feedback loop. In fact, we can revisit that content immediately and optimize it in real time, resulting in delivering the right piece of content to the right person at the right moment, through the right channel.
What are the key takeaways from this article?
- Reusing core assets will set you up to master "omnichannel" and beyond.
- It’s more than just reusing assets—it’s the process and the data, too.
- Agile operations will be important to master in almost every scenario.
The concept of agile CMS means you’ll get connections to the whole stack. You can’t operate your content without having content planning, content services, or code components that are coming from elsewhere in the technology stack. It’s no longer about trying to build something from a single tool—agile CMS represents the next chapter in the book.