
September 2023 Releases
In February, we made AI agents sound and feel more natural, even when handling complex, multi-step conversations with conditional logic.
Additional LLM insights from Regal Improve help your AI agents answer more customer questions, stay in the conversation longer, and drive better outcomes from every interaction. For multi-state agent workflows, Node-Specific Settings let you fine-tune how the agent speaks and behaves at each moment in the conversation, so transitions feel fluid, not robotic. Global Actions ensure consistent logic across every interaction, without repetitive configuration.
Together, these updates give you the control to build AI agents that don't just follow a script: they adapt, respond, and engage naturally with each customer.
.png)
We've reorganized the navigation bar into fewer, more coherent sections. With the new search bar, you can instantly find a specific contact, user, or journey, or jump to any page in the app. To find what you’re looking for, click the search icon or press Cmd+K from anywhere. This new experience respects user access permissions.
Learn More
.png)
For multi-state agents, node-specific settings let you tailor the LLM, voice, and responsiveness at the state level to match the complexity of each step of the conversation.
Instead of applying one configuration across the entire flow, you can optimize for speed and cost during simple steps like identity verification, then switch to a more advanced LLM model with a warmer voice for nuanced moments like explaining coverage or handling objections.
Learn More
.png)
As your multi-state agent handles more complex use cases, it becomes repetitive and unmanageable to configure universal actions in every node. Global Actions let you define agent logic once in the global prompt and apply it across your entire agent flow (instead of re-configuring the same logic across every node). Common use cases include human escalation ("I want to talk to a human") and opt-outs ("I need to call back later").
Learn More
.png)
With Regal Improve, your AI agent can handle every question, answer immediately, and sound like your best rep, every time. Regal Improve’s Coverage Gaps Dashboard helps you better understand not just what questions your customers are asking, but whether your AI agent is equipped to handle those questions.
The dashboard breaks coverage into three clear layers. Knowledge Base Coverage shows how well your KB content supports a topic. Prompt Coverage reveals whether your agent prompt clearly instructs the model on how to handle that topic. The new Deep Dive view lets you review real conversations with surrounding context to see exactly how the agent responded. By pinpointing whether gaps live in your knowledge, your instructions, or real-world execution, Regal Improve helps your agents resolve more issues on the first try, reducing unnecessary transfers, improving containment, and delivering the kind of seamless experience that keeps customers coming back.
Learn More
.png)
Each AI agent now has a dedicated performance dashboard with standard metrics like calls, transfer rate, containment, talk time, and sentiment, plus your unique Custom AI Analysis metrics. Filter by timeframe, add graphs, and customize the dashboard to your liking. Quickly understand what's driving performance and jump directly to relevant calls for review.
Learn More
We've released new API endpoints, so you can automate more of your Regal workflows:
Messaging
Send Message: Send SMS/MMS messages without needing to configure a journey or campaign . Learn More
Branding & Spam Remediation
→ List Business Profiles: Retrieve your Business Profiles to use when registering branded phone numbers. Learn More
→ Create Branded Phone Number: Register a phone number for spam remediation and/or branded caller ID across carriers (AT&T, T-Mobile, Verizon). Learn More
→ List Branded Phone Numbers: Check the registration and approval status of your branded phone numbers. Learn More
→ Update Branded Phone Number: Update an existing branded number's carrier registration or features. Learn More
→ Delete Branded Phone Number: Remove a phone number from branding or spam remediation. Learn More
Launch AI agents faster than ever before with Regal Copilot. Start with a short description, script or transcript to generate a first draft automatically, then iterate with AI-suggested changes, simulations, and one-click approvals.
Draft versions of AI agents let you work on and test out a new version of an AI agent, without making risky changes to your live AI agents. With drafts, you can gather feedback, refine prompts or settings, and align internally, then deploy to production only when the updated version is fully validated.
Deploy multiple variants of an AI Agent simultaneously with different prompts, voices, or settings, and control how traffic is split between them. Whether you're testing tone, objection handling, or logic paths, variants make it easy to compare performance and double down on what works faster.
Ready to see Regal in action?
Book a personalized demo.



