Admins can create instructions for the selected client. They can define multiple instructions, enable or disable them as needed, and use templates for common instructions such as tone of voice.
More information about creating, adjusting and managing instructions for admins can be found here.
Tools
Tools are what make Opal particularly powerful: they allow Opal to perform actions on your behalf. Each tool has a specific purpose, such as creating a campaign, reading campaign data or content, or generating images. Tools automatically connect your natural language request to specific actions, so you don’t need to know the underlying systems or commands. Think of tools like parts of a Swiss army knife – each piece has a very specific job.
While using Opal, you don’t need to manually call tools: Opal understands the context of your request and automatically uses the most appropriate tool. You can still mention specific tools directly in your prompt if you want to guide Opal.
Alongside system tools (available for all Optimizely products), we provide a wide range of tools specifically designed for working with Campaign. Find more information about tools and workflow agents here.
Opal credits
Opal is based on credits (similar to tokens in large language models). Every action an agent performs consumes Opal credits. The number of credits depends on the complexity of the agent and the task. Simple text tasks (like short translations) use only a small number of credits, while more complex agents consume more.
To make it easy for you to get started, all Campaign accounts on Opti ID receive:
- 200 free Opal credits per month, and
- a one-time bundle of 10,000 Opal credits, valid until 13 February 2026.
Credits are shared across products: if you use Opal with other Optimizely products in addition to Campaign, they all draw from the same credit pool.
If you have questions about credit packages or additional volume, please contact your Optimizely representative.
Find more details on how credits work and how admins can monitor usage here.
Opal in Campaign
To make it easier to get started, we’ve grouped typical use cases into multiple levels. You can think of Opal in Campaign as either an Advisor, an Assistant, or a Creator.
The first two levels are now available in Optimizely Campaign.
Opal as an Advisor
Opal acts as your creative sparring partner in chat – and can be helpful even without deep campaign data. You ask questions, and Opal responds with suggestions, explanations or analyses.
Examples of questions you can ask:
- Which subject line approaches typically work well in my industry?
- What are the current trends in B2B SaaS email marketing?
- Analyze the following competitors:
Best suited for: campaign ideas, content angles and A/B test ideas, market and competitor questions, and quick interpretation of campaign results.
Opal as an Assistant
Opal can also act as a context-aware assistant that understands and processes your Campaign data.
General analysis: “What stands out to you about the performance of our recent newsletters and how would you improve them?”
Campaign checks: Review subject lines, preheaders and copy for clarity and tone, and generate variants (Use @subject_previewtext_ideation to trigger this agent).
Campaign drafts: Ask Opal for suggestions for a welcome series or reactivation campaign that you can then implement in Campaign.
Translation & adaptation: Translate existing mailings into another language and adapt tone or target audience (You can trigger this agent directly with @translation_mail_content).
Coming soon: Opal as a Creator
We’re already working on enabling Opal to perform actions directly in Campaign on your behalf, such as creating emails, building audiences, or setting up campaign journeys. These capabilities will be delivered in a later release.
Examples: how to use Opal in your Campaign day-to-day
Here are a few examples of how you can use Opal:
Quick performance check of your latest newsletters: “Summarize the performance of our last three newsletters and give me three concrete optimization suggestions.”
Spam & quality check before sending: “Check this newsletter for potential spam risks and give me concrete improvement suggestions.”
Drafting a welcome journey: “Outline a 4-step welcome journey for new newsletter subscribers with high-level content ideas for each email.”
Email translations and tone refinement: “@translation_mail_content Translate this email into English and adapt it for a more formal B2B audience.”
Further resources & best practices
You can find all resources mentioned in this article here:
