Data Integration Guide
Find the right integration type for different data categories
This guide helps you pick the right integration option based on your data type and goal.
Quick Selection by Data Type
| Data Type / Goal | Recommended Integration Type | Start Here |
|---|---|---|
| Documents & knowledge bases | Data Pools (RAG) | Data Pools (RAG) |
| Cloud tools with live data | Connectors | Connectors |
| Internal custom APIs/services | Custom MCP | Custom MCP Server |
| Workflow orchestration | No-code (Make, n8n, Power Automate) | Make Integration |
| On-prem/private systems | On-Premise Connections | On-Premise Connections |
Typical Selection Logic
- Need live data and actions?
Start with connectors or custom MCP. - Need semantic search over large knowledge sets?
Use Data Pools (RAG). - Target system is internal/on-prem?
Plan via On-Premise Connections. - Need process automation?
Add no-code flows.
Connectors vs. Data Pools in Detail
| Feature | Connectors | Data Pools |
|---|---|---|
| Data freshness | Real-time data | Synchronized index |
| Access method | Live API calls | Vector search in index |
| Authentication | Per user (OAuth) | Centrally managed |
| Administration | Each user themselves | Admins only |
| Speed | Depends on API | Very fast |
| Offline capability | No | Yes (after sync) |
When to Choose Which Option?
Choose Connectors when you...
- Need real-time data (current emails, calendar events)
- Want to perform actions (create tickets, send emails)
- Need personalized data (only your own files)
- Want to quickly connect something without waiting for admins
Choose Data Pools when you...
- Want to search large document collections
- Want to share company-wide knowledge
- Need consistent answers across all users
- Want to make data from systems without APIs available
Activating a Data Pool
Attention
A selected data pool only becomes active after saving the assistant. Without saving, the assistant cannot access the data.
Common question: "I selected a data pool, but the assistant cannot find any documents."
This usually happens because the assistant was not saved after selecting the data pool. The configuration is only sent to the backend service when you save.
Practical Examples
Support Assistant
A support assistant needs stable knowledge and the ability to act. The combination of data pool and connector covers both:
- Data Pool: Product documentation, FAQs, known issues
- Connector: Atlassian (Jira) for creating and updating tickets
That way, the assistant finds the right knowledge base articles and files a ticket directly when needed, instead of only giving advice.
Sales Assistant
In sales, current appointments and live customer data matter more than archived knowledge — connectors alone are enough here:
- Google Calendar: Check availability for demos
- CRM Integration: Retrieve current customer data
- Email: Send follow-up emails
A data pool would mostly be dead weight here, since the relevant information keeps changing.
Legal Department Assistant
Contract templates, compliance guidelines, and internal policies rarely change and need to be searchable, not live. A data pool with these holdings is usually enough on its own — a connector adds little unless actions in a third-party system are actually needed.
Onboarding Assistant
New employees ask questions and need appointments set up for them automatically — so the combination pays off here too:
- Data Pool with onboarding materials, process descriptions, and HR policies answers the questions.
- Connector to Google Workspace sets up calendar appointments and sends email notifications.
That way, the assistant handles both knowledge sharing and part of the organizational work.
The most powerful assistants often combine both methods: data pools for stable company knowledge, connectors for current data and actions.