Designing a Custom Signal Event in Event Studio
From business problem to deployable event — a walkthrough of the Event Studio workflow
Event Studio is where you design custom Signal Events. Describe your research goal, and Savo's AI builds a complete event — dimensions and facilitation guides — that you review and publish. The process typically takes under 10 minutes from problem description to deployable event.
Accessing Event Studio
Navigate to Event Studio in the main navigation and click + Create Signal Event.
Two Ways to Start
Event Studio offers two entry paths:
Voice — Savo guides you through the intake as a spoken conversation. Speak naturally; Savo asks clarifying questions to gather what it needs. This is the default path and is the easiest way to get started if you're new to building events.
Direct Build — Type your responses in a structured intake form. Useful when you already know what you want to create and prefer working through a form at your own pace.
Both paths produce the same result and feed into the same six-stage process.
The Six Stages
Stage 1 — Intake. You describe your research goal: the business problem you're trying to understand, who the participants are, and what decisions the research will inform. An optional section lets you add further context — a website URL, background documents, and duration preferences. Click Build it to proceed.
The AI selects the interview mode based on how you've described the event. You don't choose a mode during intake — the system determines the right fit from your description. See The Four Interview Modes Explained for how each mode shapes the conversation.
Stage 2 — Researching. The AI works in the background — analyzing your problem, researching relevant frameworks and domain knowledge, and preparing the event design. This typically takes 60–90 seconds. No action is required.
Stage 3 — Reviewing Dimensions. The AI proposes the dimensions your event will measure — the specific aspects of the research topic it will explore in conversation. Each dimension has a name, definition, and an evidence basis flag showing whether it's grounded in research literature, derived from domain knowledge, or where evidence is thinner.
Review each dimension: does it map to a research question you care about? Remove irrelevant ones. Add any the AI missed. Reorder to reflect your research priorities.
Stage 4 — Designing the Flow. The AI presents the full conversation plan — facilitation guides (one per dimension) and the sequence of topics. Review the flow as a whole: does the order make sense for a natural conversation? Does the facilitation language fit your participants?
Use Edit in Event Editor for direct, field-by-field changes, or Regenerate with feedback to re-run generation with a specific correction applied ("focus more on team dynamics" or "the dimensions are too high-level for this audience").
Stage 5 — Building Spec. The AI assembles the complete event specification. This stage runs automatically after you confirm the flow — no action required.
Stage 6 — Done. Your event specification is complete. Before publishing, you can review the Research Provenance report to see the scientific and domain basis for each design decision. When you're ready, click Complete to publish the event to the Signal Event Library, where it's immediately usable in a campaign.
The Event Editor
The Event Editor is available throughout the process and after publishing. It's a multi-panel interface where you can edit every aspect of the event — settings, dimensions, and guides — field by field. Changes do not save automatically; use the Save Changes button at the bottom of the screen to apply your edits.
For detailed guidance on reviewing the AI-generated design, see Defining Dimensions and Facilitation Guides.