AI assistant (Early concept)

Product: RingCentral app (Enterprise B2B SaaS)

Platform: Desktop

Project context and goals

This project was an early conceptual exploration of an AI assistant embedded directly into the Enterprise B2B SaaS desktop app. The goal was to explore how an assistant could help users:

  • Find answers faster

  • Understand product features

  • Perform common actions on their behalf

The concept was created based on initial stakeholder and product manager requirements and was not intended as a final or fully validated solution.

Design goals

Explore how an AI assistant could:

  • Act as a conversational entry point to product functionality

  • Reduce the need to navigate complex settings

  • Translate user intent into concrete actions

  • Stay lightweight and non-intrusive

Key UX decisions

1. Progressive disclosure

The assistant does not ask all questions upfront.
Each step appears only when needed, reducing cognitive load.

2. Mixed content responses

The assistant combines text explanations, visual cards, and action buttons.

This supports different learning preferences without overwhelming the user.

3. Explicit confirmation

Before scheduling call forwarding, the assistant summarizes the date range and the destination.

This serves as a safety check for an important action.

UX approach

The assistant experience was designed around a simple principle:

Users should be able to express intent in natural language and let the system handle the complexity.

Instead of forcing users into settings screens, the assistant acts as an intermediary between user intent and system actions.

Outcome

The result was a clear conceptual direction for:

  • How an AI assistant could live inside the RingCentral desktop app

  • How conversational UX could simplify complex workflows

  • How intent-based interaction could replace manual configuration flows

While this was an early concept, it helped stakeholders visualize future interaction possibilities.

Previous

Account page