Embedded analytics concepts

Target Audience: Developers & Non technical users

TL;DR

Integrate Toucan AI visualizations and conversational assistants directly into host applications to provide contextual data exploration.

---

When to use this

Use this page to understand the architectural relationship between your application and Toucan AI, specifically regarding security, branding, and component delivery.

---

Definition of Embedded Analytics

Embedded analytics is the integration of dashboards, charts, and AI assistants into external software, SaaS platforms, or internal tools. This model provides data insights within the user's primary workflow, removing the requirement to access a standalone business intelligence (BI) tool.

---

Core Architectural Concepts

Toucan AI utilizes a "separation of concerns" architecture. The host application manages user identity and context, while Toucan AI manages data visualization and AI-driven logic.

• Web Components: Toucan AI provides embeddable web components compatible with React, Vue, and vanilla JavaScript for front-end integration.

• Security Model: Access is managed through API keys and tokens to enforce authentication and Row-Level Security (RLS).

• White-Labeling: Components can be styled to match the host application's branding to maintain a uniform user experience.

• Conversational Interactivity: End-users can interact with embedded elements through filtering or by using the AI assistant to generate charts via natural language.

---

Implementation Use Cases

Embedding Toucan AI is utilized in environments requiring secure, multi-tenant data delivery:

• SaaS Platforms: Providing customer-facing analytics dashboards as a product feature.

• Internal Tools: Surfacing operational data within corporate portals.

• Self-Service Portals: Enabling users to ask data questions through an embedded AI assistant.

---

Benefits for Developers

• Reduced Engineering Effort: Integrate advanced analytics and LLM-powered features without building custom visualization engines.

• Data Governance: Maintain centralized control over data access and security permissions across all embedded instances.

• Contextual Access: Deliver data at the point of action, reducing context switching for the end-user.

---

Constraints

• Authentication Requirement: Every embedded session requires a valid token to ensure data security.

• Network Visibility: The host application must have connectivity to Toucan AI services to render components.