Thoughts
Opal update! Key take aways from this year’s Opticon and why the AI tool is here to stay
I spent two days at Opticon 2025 in London, joining our CEO, Magnus Kerner, for product briefings and conversations with customers and partners. My biggest question going in: Is Opal just another AI assistant, or is it becoming the control plane for digital work?

What’s changed since last year?
Opal today is not the Opal from six months ago. It’s more deeply integrated across Optimizely’s stack, including the CMS (Content Management System), EXP (Experimentation), and CMP (Content Marketing Platform), and handles tasks such as creating campaigns from conversational briefs and generating page types directly from Figma designs. The velocity of integration stands out.
Beyond another chatbot
The most important shift is architectural: Opal behaves like an orchestration layer, not just a chat interface. In Optimizely’s framing, Opal (agent orchestration) sits above Systems of Insight (analytics, RAG) and Systems of Action (CMS/EXP/CMP), turning intent into coordinated actions.
Contextnative by design (“Opal knows”)
General-purpose models require you to restate the context for every task. Inside Optimizely, Opal already has the context, your content, assets, brand data, analytics, and documentation, through retrieval-augmented grounding. That drastically reduces setup friction and shifts effort to decision-making, not prompt plumbing.
What this unlocks in practice
Agentic workflows can run continuously: digesting new inputs, proposing experiments, generating variations, checking compliance, and updating assets. The result isn’t just speed; it’s surface area - more tests run, more issues caught earlier, more opportunities unearthed.
Extending to your environment
Out-of-the-box tools cover common marketing and digital ops tasks, and we can build custom tools so Opal can act on your specific systems and data. That’s critical for moving from demos to durable workflows.
So, is the AI bubble about to burst?
Yes, for thin wrappers on generic models. Those will fade. However, platforms that combine orchestration, first-party context, and instrumented outcomes won’t. Opal is heading in that direction. The next milestone will be evidence, including adoption metrics and outcome deltas, which we will share as concrete numbers become available. We are looking forward to helping customers operationalize Opal in their DXP.
Takeaways
- A new operating model is required. Working with Optimizely now means adopting a journey-centric, continuously orchestrated way of working. Success hinges on change management and new skills (content ops, experimentation, agent/tool design) for both customers and partners.
- New operating model: From project-based CMS builds to AI-orchestrated marketing journeys. Less heavy engineering, more configuration and experimentation owned by marketers, underpinned by structured change management and competency uplift across teams and partner ecosystems.
- Implication for teams & budgets: Plan for a rebalancing from bespoke builds to orchestration and enablement as Opal automates routine production (e.g. campaign creation and Figma-to-page-type flows). Invest in training and partner capabilities around journey design, AI tooling, and data governance to realize value.
Interested in exploring Opal?
We help teams design agentic workflows, integrate with CMS/EXP/CMP, and measure lift. Contact us for a technical working session!
Disclaimer: Gartner is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally, and Magic Quadrant is a registered trademark of Gartner, Inc. and/or its affiliates. The report is Gartner’s copyrighted work and is used herein with attribution.