Wobby Explorer

Wobby Explorer is the interface where business teams get answers from their data using AI agents. It's designed for non-technical users who need insights without writing SQL or understanding database structures.

Overview

Explorer provides a conversational interface to pre-configured AI agents. Business users ask questions in natural language and receive answers as summaries, tables, and visualizations - all backed by actual queries against your data warehouse.

While Wobby Studio is where data teams build and configure agents, Explorer is where the value is realized. It abstracts away the complexity of data infrastructure, making self-service analytics accessible to everyone in the organization.

Who Uses What

Access Control

Wobby uses role-based access to separate technical configuration from daily usage:

• Studio role: Full access to both Wobby Studio and Explorer. For data teams who need to configure agents and test them.

• Explorer role: Access to Explorer only. For business users who interact with configured agents.

Mobile Support

Explorer is fully supported on mobile devices, allowing users to query data and view results from anywhere. Studio is desktop-only due to its complex configuration interface.

Key Features

Conversational Follow-ups

Unlike traditional BI tools, Explorer maintains context throughout your analysis session. After receiving an answer, you can:

• Highlight any text in the response to ask for clarification or dive deeper

• Reply with follow-up questions that build on the previous context

• Explore related metrics without starting from scratch

This creates a natural analytical flow - start broad, then drill into specific areas of interest as patterns emerge.

Analysis Modes

Explorer offers two distinct approaches to handle different types of questions:

• Quick Analysis: Direct queries that return single answers in under a minute. Perfect for KPIs, specific metrics, and routine checks.

• Deep Analysis: Multi-step research for complex questions. The agent creates a research plan, executes multiple analyses, and synthesizes findings into a comprehensive report.

You can read more about the difference between the two in our dedicated documentation.

Collaboration & Debugging

Every analysis in Explorer generates a unique URL that can be shared with teammates. This ensures everyone sees the same data and conclusions. The data remains protected and is only accessible to invited users.

For data teams, the Debug in Studio button provides direct access to the full execution trace - invaluable when business users report unexpected results or need help understanding how an answer was generated.

Suggestions

Suggestions are curated starting points that help users understand what they can ask and discover insights they might not have considered.

How Suggestions Work

Suggested questions appear organized by category (e.g., "Customer Analytics", "Revenue Metrics"). Users can click any suggestion to run it immediately, making it easy to explore data without knowing exactly what to ask.

These suggestions serve two purposes:

• Onboarding new users by demonstrating the agent's capabilities

• Surfacing important metrics that might otherwise go unnoticed

Managing Suggestions (Data Teams)

Data teams control suggestions through Wobby Studio. Navigate to your agent and open the Suggestions tab.

Creating Suggestions

Wobby can auto-generate suggestions based on your data structure and metadata:

1. Click "Re-generate All" to create suggestions automatically

2. Review the generated questions

3. Add your own custom questions for business-critical metrics

4. Edit any suggestion to match your terminology

Protecting Suggestions with Pins

When you're happy with a suggestion, pin it to protect it from being overwritten:

• Click the pin icon next to any suggestion you want to keep

• Pinned suggestions survive when you click "Re-generate All"

• Unpinned suggestions get replaced with fresh auto-generated ones

This lets you keep your carefully worded KPI questions while still discovering new analytical possibilities through regeneration.

Best Practices

• Pin all manually created questions

• Pin auto-generated questions that perfectly capture important metrics

• Leave exploratory questions unpinned so they refresh with new patterns

• Organize categories to match how your business thinks about data