Optimizely Opal overview

  • Updated

Optimizely Opal is an agent orchestration platform that helps you work smarter across Optimizely One. Whether you create content, manage experiments, or analyze data, you can use Opal to automate tasks, surface insights, and guide decision-making. Opal has a flexible system of AI agents that are intelligent software components that understand your intent and help you reach your goal.

See Optimizely Opal Essentials in Optimizely Academy for more information.

AI background

AI refers to software that performs tasks typically requiring human reasoning, such as learning from data, making decisions, or solving problems. These systems identify patterns, generate content, and offer insights using large language models (LLMs) and other algorithms trained on vast datasets.

Recent advances in computing power, data availability, and model design accelerate how people use AI across industries. At Optimizely, AI is embedded into the platform to help you move faster, automate repetitive work, and make more informed decisions.

AI agents

AI agents are software systems that act on your behalf. They can analyze information, interact with the platform or other tools, and adapt their behavior based on feedback. Unlike traditional tools that only respond to direct input, agents can manage goals and take steps to accomplish them.

Optimizely classifies AI agents into the following three broad types:

  • Simple assistants – Help with straightforward tasks such as generating content, formatting it into a structure, or offering ideas based on a prompt.
  • Specialized agents – Focus on a domain or tool. These agents use expert-level context to analyze data, predict outcomes, or recommend next steps.
  • Workflow and autonomous agents – Coordinate and perform multi-step processes, often by integrating with other tools. These agents can handle tasks across systems and adapt workflows over time.

These agents improve productivity by reducing manual work, increasing accuracy, and providing a tailored experiences. They also support scalability, helping teams achieve more without growing headcount.

Optimizely Opal

Optimizely Opal is your AI assistant across Optimizely One. It responds to natural language questions, generates insights, and lets you complete tasks faster. 

Optimizely Opal uses AI agents to 

  • Answer platform questions – Understand terminology, workflows, and configuration steps.
  • Configure content, campaigns, and experiments – Save time during creation and planning.
  • Automate repetitive tasks – Reduce manual configuration and speed up delivery.

Optimizely Opal uses Opal credits.

What you can do with Opal 

Most interactions happen through Opal Chat, which is available across the Optimizely platform in the global navigation bar. 

image.png

Whether you are launching a campaign or fine-tuning an experiment, Opal adapts to your role and needs. Use natural language to ask questions, generate content, or request help. Opal responds in real-time and remembers threads so you can pick up where you left off.

The following are some examples of what Optimizely Opal can help with:

  • Use tools – Ask Opal to complete tasks using built-in tools, like creating a campaign, drafting content, analyzing images, or suggesting flag variations. See Tools overview.
  • Analyze files – Upload a document, image, or spreadsheet and ask questions about it. See Optimizely Opal Chat.
  • Answer questions – Ask Opal about Optimizely features, capabilities, or concepts. It can explain terms, walk you through configuration steps, or point you to relevant documentation.

See Products with Opal features for a list of all available Opal features.

How Opal works

When you ask Opal a question, it follows a structured process behind the scenes.

  1. User prompt – You enter a request in natural language. Your prompt can be a simple question, a complex task, or a creative brief. Opal analyzes your intent, extracting key information to guide its next steps.
  2. Instructions – These foundational guidelines define Opal's persona, approach, tool usage, evaluation criteria, and response formatting, influencing every step of its operation.
  3. Contextual intelligence – Opal adds context using content knowledge, Optimizely platform data, and your workspace context. This ensures Opal's responses are informed by your specific environment and goals. 
  4. Strategic tool selection – Opal selects the right tools for your task based on your request and the enriched context. Tools let Opal perform a wide range of tasks from data analysis to content generation.

  5. Agent input generation and validation – Opal needs specific pieces of information to execute any tool or agent. These pieces of information are often called 'parameters' or 'arguments'. Think of them like the specific fields you fill out in a form, or the instructions you give a smart home device, like a smart thermostat. Just like the smart thermostat needs to what temperature you want the house set at and at what time it should turn on and off, Opal needs these specific inputs to understand exactly what you want it to do and to execute the task accurately.

    The following are example agent input parameters:

    • A search_query when you ask Opal to "search the web for marketing trends". 
    • The recipient_email_addresses, subject, and body when you ask Opal to "send an email". 
    • A file_key or url when you ask Opal to analyze a design. 
    • A start_date and end_date for a report.

    Opal validates these inputs to ensure they are correct and complete. If any crucial information (a 'required parameter') is missing or invalid, for example, if you tell the smart thermostat to set the temperature to "banana", Opal asks you for clarification. This ensures the command is understood and can be executed successfully.

    • If the agent input is valid – Opal proceeds to the LLM interaction.
    • If the agent input is invalid – Opal goes directly to the Execute and refine (step seven) to request clarification from you. This prevents unnecessary LLM engagement.
  6. LLM interaction – Opal leverages advanced AI models to process complex information, generate creative content, and derive actionable insights. This is where raw data transforms into intelligent solutions. Opal uses foundational LLM models through business accounts, so your data is never used to train the model or shared across customers. 
  7. Execute and refine – Opal executes the planned actions (calling agents and tools as needed), and iterates. If the initial output is not perfect, Opal can re-evaluate. This continuous feedback loop ensures optional results and measurable impact. If clarification is needed, Opal asks you for more information.
  8. Tailored response – Opal delivers a clear, concise, and actionable response. The response is often presented within the Opal Canvas or user interface, where you can interact with and manage generated content or executed actions.

Why use Opal and AI agents?

Opal helps you with the following:

  • Work faster – Eliminate repetitive tasks and reduce manual configuration.
  • Scale content and insights – Get more done with fewer handoffs.
  • Improve personalization – Tailor output based on your brand, audience, and goals.
  • Make informed decisions – Use data-backed recommendations without digging through dashboards.
  • Simplify complexity – Let Opal manage workflows so you can focus on strategy.

See Optimizely Opal and AI features for a list of all available Opal, generative AI, and machine learning features.

Products with Opal features

Administrators must grant users access to Optimizely Opal.

Opal is enabled by default for eligible customers, but users do not have access until an administrator enables it for each individual using Opti ID. For steps, see Get started with Optimizely Opal for Admins.

Optimizely Opal app

Opal administrators can extend Opal's base capabilities and

Opal users and administrators can

Optimizely Analytics

Campaign

Collaboration

  • Use Opal to write experimentation plans and suggest hypotheses and variations.
  • Use AI to refine variations and written briefs.

Commerce Connect

Configured Commerce

Content Marketing Platform

Content Management System (PaaS)

Content Management System (SaaS)

  • Chat with Opal for CMS (SaaS).
  • Create content models automatically by having Opal analyze URLs or images using the Content Model Creation agent.
  • Identify, sort by last edit date, and help resolve duplicate and outdated content (in terms of content guidelines as reference) across a website in CMS (SaaS) using the Content Refresh Analysis agent.
  • Enhance your CMS (SaaS) content's visibility and (LLM) discoverability by automatically identifying opportunities for structured data markup with the GEO Schema Optimization agent.
  • Evaluate existing SEO properties, identify optimization opportunities, recommend improvements, and upon your approval, create or update SEO metadata fields in your CMS (SaaS) instance with the SEO Metadata Implementation agent.

Content Recommendations

  • Use the system tools to retrieve the highest-performing content items, gather performance metrics for the most engaging topics, and analyze the performance of the recommendation widget.

Feature Experimentation

Opti ID Admin Center

Optimizely Data Platform (ODP)

Optimizely Opal is available only to United States–based ODP customers.

ODP + Optimizely Content Recommendations

Optimizely Opal is available only to United States–based ODP + Content Recommendations customers.

Personalization

Product Information Management (PIM)

Web Experimentation

 

If you use Opti ID, administrators can turn off generative AI in the Opti ID Admin Center. See Turn generative AI off across Optimizely applications.

Next steps