A good conversation often begins with food. Or, at the very least, a hearty discussion about food.
That’s exactly where my recent chat with Darren Guarnaccia began, reminiscing about the darkly hallowed back room at El Gaucho in Seattle, where we dined together after an AWS shindig in 2019.
At the time, Darren was at Crownpeak, an enterprise CMS vendor in Amazon’s burgeoning DCX Competency program (BTW, that’s an acronym for Digital Customer Experience, not to be confused with DXC).
If you’re unfamiliar with this storied establishment, El Gaucho is a swanky, dimly lit steakhouse with dark wood and nostalgic fixtures. Opened in 1953, the Northwestern chain conjures the elegance of a bygone era that starkly contrasts the hip tech culture synonymous with the Emerald City.
The food, you ask? Let’s just say that the chateaubriand will make your toes curl, and the wine is divine. You might need a candle to find the utensils, but it’s worth the effort.
As we pieced together the remnants of that dinner, it occurred to me that El Gaucho is a metaphor for CMS. Much like a well-aged steak, the content management system is a classic fixture on the digital menu. And while the choices have evolved (I’ll have mine medium headless, thank you) it’s still an essential course.
But appetites changed. We shifted to a DXP “buffet” of services that included (almost) everything in the digital stack. In short order, enterprise CMSes gobbled up marketing automation systems, CDPs, personalization engines… the whole monolithic menu.
In lockstep, Gartner retired its web content management MQ, focusing instead on DXPs and their completeness of vision in areas like data, testing, and optimization. In this new omnichannel world, websites were just one dimension of customer experience – and CMS was becoming another layer in a complex stew of interconnected tools.
Then came composable. And maybe… the death of the DXP?
In Darren’s new role as the President of Uniform – a trailblazer in digital experience composition (DXC) – he’s leveraging decades of tech experience to chart the course in this nascent space. But therein lies the real challenge: building a category that has yet to be blessed by analysts. It's worth noting that the freshly-minted 2023 Gartner Magic Quadrant for DXPs was just released, and DXC was mentioned as an important industry trend.
If there’s anyone uniquely designed to guide Uniform to the next level, it’s Darren. The man is tenacious, piloting absolute command of his message with an unflappable focus. He’s a proven force in the DX community, leveraging a fluency that spans the stack and experience that builds off his tenure with industry stalwarts like Sitecore.
All that, and he has great taste in food.
What it amounts to is a killer combination in a leader – and exactly what Uniform needs if it plans to grow DXC and drive the composable movement.
Finding a perfect fit at Uniform
Still fresh to the reins, I asked Darren what attracted him to the opportunity at Uniform. Previously, he held the roles of CPO, CMO, and Visionary Product Strategist at HootSuite, a highly adopted social media management platform.
“I love emerging markets like this,” he said. “To me, this DXC space is so interesting because we’re back to a tectonic shift in technology. There’s this huge gap in the market, particularly for underrepresented user bases. I love being able to come to the rescue and solve these problems – and really rescue the user, as I like to call it.”
This user-centricity harkens back to his role as the first product marketer at Sitecore in 2007, where he got hooked on customer experience. The company was one of the earliest to champion this idea and develop a novel customer engagement platform. While the tools evolved to serve marketers in the quest for CX, the rise of omnichannel and more complex, API-driven strategies have swung the pendulum back to the IT department.
“We’ve done a great job in headless for the devs,” he said. “They’ve got this amazing freedom of choice. But what the hell happened to the marketing team? How did we get back to this place where the IT team is specifying which technologies a marketer has to use? That just feels like a whiplash moment – and for me, it seems like a wrong that needs to be righted. I get very passionate about that.”
Darren brings that passion to Uniform as it looks to establish a foothold in the stack. A bit of backstory: the company was founded in 2020 by Sitecore veteran Lars Petersen and rapidly moved from blueprint to business, raising $28 million in a Series A investment just 15 months after launch. This capital put serious wind in the company’s sails, allowing them to invest in the right people and technologies to scale.
For as long as I can remember, Lars has been jockeying from the Jamstack stable, leading the composable race in an API-first world. I even recall the prototype of Uniform making an early debut at one of our CMS Experts group meetings hosted by Janus Boye. At the time, the DXC moniker was still years from conception.
Despite the gravitas that Lars and Darren bring to this emerging space, one of the biggest challenges to scaling is a general lack of awareness – and even confusion – about what a DXC is.
As defined through the Uniform lens: DXC is a combination of technology and practices that gives developers more control over deploying composable architectures without having to manually create expensive and custom glue code for each service (more on glue code in a bit).
Uniform’s DXCP promises users a faster path for connecting CMS, commerce, DAM, search, and even legacy services simply by adding API keys to its Uniform Mesh. What’s more, DXC provides marketers and business users total control over the look and content of their experiences across every channel while leveraging CDN integrations to power A/B testing and personalization.
In Darren's words: “We're more like a Zapier than anything else, because we're really just a translation layer of back end to front end with some visual tooling to let you assemble elements.”
I probed around this continued challenge of awareness, and he reinforced that it’s still early days – and education is key to building demand.
“It feels a little bit like mobile was,” he said. “When it happens, it's going to require a whole new way of delivering stuff. And I think that's when that's when headless and composable [will] become really interesting.”
Will DXC replace DXP?
In January at NRF’s big retail show in New York, the MACH Alliance hosted a series of composable chats under the banner of MACH Haus. The panels featured numerous MACH-focused sessions, including an undeniable entry titled “The Death of DXP – Stepping Off the Monolith Bit by Bit.”
Darren was among a cabal of experts on this charged topic, along with John Torris of Bold Commerce, Karen Light of Valtech, and Chris Bach of Netlify. CMS Critic contributor Sana Remekie of Conscia.ai moderated the session.
The Death of DXP. Provocative? Sure. But I remember reading headlines in early 2020 about Gartner’s decision to “kill” the web content management category, and how this might signal the end of CMS as we know it. The consensus from Gartner was less sour, suggesting WCM had fully matured and content was transforming with omnichannel.
So less headstone... more retirement plaque? I know a few DXPs – some calling themselves composable – that aren't betting on either scenario.
I asked Darren about a question he posed in a recent LinkedIn thread to Irina Guseva, lead analyst of the former WCM Magic Quadrant at Gartner. His query to her was whether DXC would replace DXP.
Is this really the evolutionary trajectory? Will DXP users flock to DXC and go composable?
“I think the real question is why do you move,” he said. “Why do people come to headless? Why go composable? What's the driver behind all that? I think it's a combination: there's definitely the developer frustration story of ‘I'm tired of being handcuffed with whatever platform,’ whether it's an Adobe, Drupal, Sitecore. Some want that developer flexibility and freedom. For others, it’s a velocity problem. It's [also] expensive to make these things scale.”
He also noted that performance is becoming a critical factor, citing how good Lighthouse scores and SEO are becoming increasingly difficult to achieve and maintain. And of course, there's the persistent challenge of legacy tech and bending monoliths to your will.
“You can try and get their ‘flavor’ of it right. But if you’re trying to use your framework of choice – and you want that – then you’re now thinking real hard about headless and composable.”
Okay. But will DXC supplant DXP? Will these API-driven orchestrators sunset DXP products as we know them and drive organizations to build their own MACHified stacks?
“We’re not going to convince people to leave DXP,” Darren countered. “You're going to leave on your own. But we’ll help you make that transition once you've made that choice. And you make that choice for your own reasons – like performance.”
Slaying the “Glue Code Monster”
During our conversation, Darren shared a composability graph outlining the architectural choices for building digital experiences. It acts as a sort of keystone for translating MACH maturity and as it relates to the overall composability index of a solution:
As Darren explained, the axes define the composable scale from closed to open systems. And while some DXPs might enable pieces of their stack to be leveraged in an API-driven manner, they still act as a “walled garden” where you lose capabilities the minute you pull out.
“In a pure composable world, you should be able to replug, unplug, and add without writing code – and without losing functionality,” he said. “The legacy vendors just haven't done any of this. They're the monoliths.”
If there’s one theme that keeps popping up in my conversations with vendors, it’s the notion that going composable could breed a new monolith of sorts. I spoke to Darren about the underlying factors that contribute to this potential “Frankensteining” into a MACH monolith.
“The dirty secret in the industry is the amount of code you write,” he said emphatically. “We've been talking about glue code and playing with this idea of a ‘Glue Code Monster’ to personify the problem.”
This monster is more than a clever concept – it’s a campaign with legs. I saw Lars Petersen wearing his Glue Code Monster t-shirt on a LinkedIn webinar earlier that day, and it works. The motif is fun, relatable, and crystallizes the root pain point. I asked him how it’s being received, and he said most people and agencies get it because they’re doing many of these types of projects.
“But if you haven't done five or six of these things, you haven't seen the pattern yet,” he clarified. “You're about to paint yourself into a corner with glue code and you've just created a new monolith.”
The phrase Darren invoked around this phenomenon is you've now composed yourself. It’s a sticky situation where a composable solution has become immovable because the glue has dried.
“Now you're stuck again,” he punctuated. And this is why the mindset is equally as important as the tools and technologies.
In my last interview with Casper Rasmussen, senior VP at Valtech, we spoke at length about the core principles of MACH and why adopting best practices is key to success. In many ways, this parallels what Preston So elegantly outlined about headless a few years back and why thoughtful planning is so critical to the ecosystem.
Glue code is yet another challenge to realizing the benefits of a composable solution, and it’s clear that Uniform is positioning DXC as an answer to slaying the monster.
Where do agencies and SIs fit in?
Okay, so less glue code means fewer headaches. More scalability. Greater composability. Plus, we’re giving marketers and business operations the capacity to be more agile and orchestrate omnichannel experiences in an API world.
But what’s in it for the agencies and SIs that subsist on building expensive solutions with lots of custom code? What incentive is there for them to embrace DXC as a new technology when it might impact revenue?
“Integration code is the lowest value code,” Darren said. “It’s tech debt to bail. On the one hand, you can sell a nice retainer to manage that, but it's low-value work, right? So, do you want to have a lot of those services, or would you rather have more strategic services?”
This makes sense, and from Darren’s perspective, most agencies aspire to create value-based retainers that deliver long-term revenue based on performance. He said there's a shared risk model where SIs and their customers focus on demonstrable business impact through faster iterations and cycle times.
“If I were running an agency or SI right now, I would be using these things to speed up time to value because that's how I'm going to get the follow-on engagements,” he said. “That's how I'm going to build momentum with my client and create longer duration kinds of engagements.”
Demand for DXC is growing
For anyone straddled by the yoke of enterprise software sales, there’s an obvious question about any new technology: can I actually sell this stuff?
I’ve heard from several voices in the community that demand is beginning to percolate in substantive ways. I asked Darren what he’s been seeing at Uniform and if the rumors were true.
“We're now seeing DXC RFPs,” he confirmed. “Some of them start with a general RFP, or they go buy a headless CMS first and ask, 'Well, what do we do? How are we going to assemble this stuff? Who's going to make these things?'”
He went further and talked about DXC as a salve to what he calls “backlog hell” for marketers. Even when headless is purchased for a project, users struggle with the technical implications, and they’re seeking out a solution. This is where DXC is realizing more opportunities.
“Everything's a developer ticket,” he said. “And [as a marketer] I'm waiting, you know, two, three, four sprints. How am I going to go fix this? So part of it’s like, you don't know what you don't know. You get into this headless stuff thinking it's going to make me faster. We're going to get more agility. But it doesn't actually lead to any agility.”
What’s next on the DXC roadmap?
In a recent webinar, Kelly Goestsch – the Chief Strategy Officer at commercetools and co-founder of the MACH Alliance – talked about the high cost of doing business with lots of vendors. This supported his earlier observation on LinkedIn, where he proclaimed that composable isn’t for everybody. That’s why Shopify exists.
Darren affirmed this, suggesting that DXPs persist in part because they offer a panacea to the pain of buying from multiple vendors. When it comes to problem resolution, they also provide one throat to choke.
“Overcoming that friction in the buying process is a worthy thing to tackle. That’s the other advantage a suite like Adobe and Sitecore have: you’re buying one vendor in one simple way.”
Given the ground we covered on legacy DXPs and the new breed of MACH vendors, I asked Darren if he had any thoughts on the role of the MACH Alliance (of which Uniform is a member) as a mechanism for maintaining standards. My question comes at a moment when some headless platforms are expanding into areas like hosting to grow revenue.
“I think it’s ironic that a lot of headless vendors are becoming MACHless themselves,” he said. “At some point, should the MACH Alliance keep vendors in their swim lane or from becoming too much of a monolith? That’s a very interesting question because I think the reason MACH was created in the first place was to overcome the Adobes and the Sitecores of the world.”
When I asked Darren what’s next for DXP and DXC, he first referenced Joe Cicman, an analyst at Forrester, who said: Digital experience platforms (DXPs) exist, but they’re built by enterprises, not bought from vendors — yours is no exception.
“I think this is true of what the vendor builds, and equally the end solution where you might buy a bunch of composable pieces,” Darren codified. “But if you paint them together with glue code that's composed – it's a monolith. So now you're down in the bottom right [of the index], even though the elements of your implementation are not.”
Darren reinforced the need to bring back all the useful stuff that people relied on, especially marketers. He also believes that DXC is more than a product – it’s a new way for modern teams to work faster and create content in a waterfall manner.
“I call it improving time to learning,” he said. “If what marketers really want to do is iterate quicker, out-learn, and out-iterate their competition, then how do we support that?”
Before wrapping up, I returned to Darren’s previous comments about the nature of true composability to re-plug, unplug, and add components without writing code. I asked him if this same requirement ought to be a fundamental trait of a DXC itself – in essence, engineering the "swapability" of Uniform in a composable architecture (or transplanting the “brain,” as Sana Remekie would contend).
“It should be easily replaceable,” he said. “I think it's more replaceable than most [components] because it's designed to be loosely coupled. So at worst, you export your configurations and your mappings because that's really what you do with Uniform: you map and source systems to content objects. You're not writing code and transformation to do that.”
Because that would be glue code. And that’s where things get sticky.
Are we seeing the beginning of the end of DXP? As Darren pointed out, we’re still crossing the chasm. But when it comes to the future of composable, he’s a believer – someone with the right mix of vision and experience to drive the mission forward and a passion for solving big problems in an undiscovered country.
Interviews don’t get much better than this.
Except, of course, when there’s chateaubriand involved.