XM cloud and the future of Sitecore: A few questions answered

Edward Kapuscinski

XM cloud and the future of Sitecore: A few questions answered

I had a wonderful conversation with one of our Fortune 500 clients about what they should do with their current Sitecore investment, given the current state of Sitecore's product mix. The conversation started with "What do I need to know?"

I realized that if this client has these questions, many other folks likely do, too. So here are some quick answers to a few common Sitecore customer questions about XM Cloud and the future of Sitecore.

"When Do We Need to Get to the Cloud?"

This customer hosts Sitecore 9 on-premises, so a move to the cloud would be a large project — and a big paradigm shift for their internal stakeholders. The answer to this is "Not for at least 6 years, if then.”

Sitecore not only recently released Sitecore 10.3, but they have also indicated that there will be a 10.4 sometime next year. Following the standard Sitecore product support timelines, there should be a nice long runway ahead for on-prem Sitecore XP solutions.

"Is Headless a Requirement?"

My answer here was "Only if you want to move to XM Cloud." As long as they're hosting Sitecore themselves, either on-prem or on VMs with a cloud provider (like AWS or Azure), they’re perfectly fine keeping their .NET-based solution (even using my old favorite, Webforms). This is true for all Sitecore customers.

That said, if they do want to move to XM Cloud, they’ll need to move into a JSS-powered headless architecture using React, Angular, or Next.JS.

"Why Would I Want to Go to XM Cloud?"

My answer here was "If you want to improve your site performance using a modern front-end framework, it's the best path." I’m not a fan of adopting new technologies "for the heck of it" and only really recommend that our clients do so when they bring real value. This was very much my feeling on "headless" sites.

We had a perfectly functional solution using .NET MVC to generate and deliver website experiences, so why take on the risk of something new? I found the answer to that when doing a tutorial on Next.JS: Modern frameworks, especially Next, are much more "fit for purpose" for delivering websites. Unlike previous paradigms, where the communication between user behavior on a loaded page wasn't "organic" to the technology serving those pages, Next manages the entire lifecycle of browsing. This means that pages and parts of pages can be rendered at the ideal times to optimize interactivity and performance.

My mind was blown when I learned that Next will pre-render content based on what links are visible to a user, so when they click a link, the content is immediately available. Wow! That was the moment the lightbulb went on and I realized why people would want to adopt this new paradigm.

If you're thinking about building a headless experience using Sitecore JSS, you might as well do it on the best infrastructure available, and that means XM Cloud.

“What Are the Benefits of the SaaS Composable Model?”

The big benefits are centered around environment flexibility, the ability to easily scale and provision new services on the fly, and, most importantly, have an integrated content delivery and CDN system where your entire website lives "at the edge" (if you use Next.JS and Vercel).

Our conversation centered on weighing those advantages against the risks of adopting an unproven model (at least in this client's case) and having to manage the change to an entirely new development paradigm. While I'm a believer in the benefits of the new world, the risks of the latter aspect should not be ignored.

We've recently been helping a client with some intermittent production outages that are vastly complicated because there is no server to log into and monitor the performance of, no IIS logs to pull up in the log viewer, and all the old knobs and levers that experienced Sitecore folks are used to being able to mess with aren't there. There are new equivalents, but there's a learning curve for everyone involved, even Sitecore experts.

Sadly, the only way to get experience with this type of troubleshooting is for clients to have trouble, and there's been a reset button pushed on what those troubles are.

"What Do I Need to Do to Adopt a Headless Composable Sitecore Solution?"

This question was about the actual steps that would need to be taken and the new skillsets the client's team would need to develop. The first part was easy: Rebuild your site using a modern front-end framework connected to Sitecore JSS.

I recommended Next.JS because it’s the framework Sitecore seems to be recommending. This is required because XM Cloud does not work with .NET MVC solutions. The only way to connect to the CMS is headlessly, so sites powered by it MUST connect via JSS. If they wanted to, they could use React, Angular, or even .NET Core, but whatever they do, they'll have to rebuild the presentation layer, which is, of course, a big project. The ROI on it would need to be thoroughly analyzed, but a few strategies could help make those calculations more favorable.

First, don't just "do it to do it." Key the rebuild to a larger initiative, something like a major brand or website refresh, a change in branding, or a company acquisition. Those situations often larger projects, and in those cases, the additional effort required to redevelop things that already exist would likely be significantly lower. Credera's team is also working on a potential solution for speeding the conversion up as well, but it's still too early to talk about the details.

The second part of that question is about skill-building: What new skills will developers need in the new paradigm? The answer here, first and foremost, is knowledge of whatever modern framework the JSS-powered site will be in. Whatever you’re going to use — React, Angular, Next.JS or .NET Core — now is the time to start getting your team trained in it.

Of course, learning the ins and outs of Sitecore JSS will also be important, since it can slightly change the development workflows of the tools being used. Sadly, Sitecore's training on these new paradigms hasn't quite caught up yet, but I'm sure it will soon. When that time comes, I recommend that developers of all skill levels jump on it.

A Final Thought

The last thing I left them with was that, since they are still on Sitecore 9, it’s absolutely time to upgrade to 10.3. Not just because they are now out of mainstream support, but also because 10.3 will allow them to start building the technical bridge to a potential XM Cloud future.

By working with Credera, they have access to our Sitecore Elevator tool, which, much to the delight of their lead developer, is going to make the upgrade process significantly less painful than last time they had to do one.

I think this conversation was very beneficial for our client, who now knows "what's up with Sitecore.” Here's hoping it will be for you as well.