In early May 2025, with the launch of the newly evolved version of Optimizely Opal, Optimizely marked a moment that goes far beyond a product announcement for the entire CMS and DXP market. On May 7, 2025, Optimizely presented the next evolutionary stage of Opal — with a completely redesigned interface embedded directly in the entire Optimizely One product suite, empowering marketing and digital teams from task automation and brand-compliant content creation through to end-to-end experimentation analysis. But what is truly significant is not the feature set. It is the paradigm behind it.
What happened?
Optimizely Opal is based on Google's Gemini models and thus relies on a deep partnership with hyperscalers to integrate AI capabilities directly into the content lifecycle. The system builds so-called “industry-leading context windows” by combining diversified data sources from the company's unique know-how — existing assets, campaign performance and experimentation results — and has an extensive memory enriched by Optimizely's domain expertise. In short: Opal knows who your company is, what you have done so far and what makes sense next.
Since the official launch in May 2025, nearly 900 companies have adopted Opal, executing almost 10,000 AI-powered marketing actions daily in more than 50 countries. This is not a lab observation — this is market validation in real time.
Why is this technically and strategically significant?
This launch is not an isolated event but part of a fundamental shift that the Gartner DXP Magic Quadrant 2025 described with rare clarity: Gartner predicts that by 2026, at least 70% of organizations will be required to acquire composable DXP technology instead of monolithic DXP suites — compared to 50% in 2023. Composability is therefore no longer a differentiating option but a basic prerequisite.
As AI-driven discovery and automation expand, the quality, structure and governance of content and data become the decisive success factors. Agentic AI can only be as good as the content and data it can reliably access — which makes disciplined content coordination and data hygiene the critical foundation.
Herein lies the core of the strategic significance: AI agents in the DXP context are not content generation buttons. AI in digital experience management has evolved from simple content generation toward orchestrating entire marketing workflows — agentic capabilities are becoming the new competitive battlefield. Optimizely positions Opal not simply as something embedded in a DXP, but as a potential orchestration layer that can operate above or even across platforms. AI agents become the interface, and the system that coordinates them gains strategic power.
This is the moment the industry has discussed theoretically for years: the platform is undergoing a fundamental shift — from AI as a feature to AI as the operating layer for modern marketing teams.
What does this mean concretely for companies and projects?
From a consulting perspective, this results in three direct implications:
First, content governance becomes a core technical task. Anyone with unstructured, poorly tagged or inconsistent content will not benefit from agentic AI — on the contrary, bad data will be scaled at high speed by agents. As AI-driven discovery and automation increase, the quality, structure and governance of content and data become the limiting factors for success. Agentic AI can only be as good as the content it can reliably access.
Second, the vendor selection logic changes fundamentally. The DXP market has reached a point of broad capability parity. Nearly every serious vendor offers content management, customer data, personalization, automation and commerce integration. What distinguishes leaders from laggards is no longer the presence of features, but how intelligently these capabilities are orchestrated to achieve measurable business results.
Third, “agentwashing” is the new greenwashing. The most common mistake is to call AI assistants agents — a misconception known as “agentwashing”. Companies must learn to distinguish between an AI feature in the UI and genuine autonomous agents that execute multi-step workflows without human intervention. The numerous demos of 2025 have made it clear that agentic AI holds great potential — but in production use, successful companies will need disciplined engineering and governance to unlock the real potential.
My clear assessment: The market is at a crossroads. Gartner analyst Irina Guseva put it pointedly: “DXP as we know it will cease to exist in the next few years — or even in a few months, depending on the digital maturity of the end-user organization.” This is not a threat to the industry — it is an opportunity. But only for companies that act now: systematize content quality, consolidate composability architectures and treat agentic AI strategies not as a future topic but as an ongoing project. Those who wait will not become slow — they will become irrelevant.
