---
title: "How we built a personal data scientist for every Notion employee"
description: "Notion built Data Scout, an internal data agent connecting Snowflake and Notion, letting employees query governed data in plain language and cutting request turnaround from days to minutes. Within two months, weekly data team requests fell by about 80 percent, and roughly half of all employees now use it. The agent enforces Snowflake permissions per user and uses Notion business context, enabling teams such as Sales, Engineering, and Customer Experience to ask questions about account adoption, onboarding drop-off, or support volume using shared definitions. Using MCP, teams built Data Scout Reporter for scheduled reports and domain-specific agents that reuse the same context. Notion learned correct SQL could still answer the wrong question due to four failure modes, so they built governed access, curated context, instructions, query tracing, and a review process where change evidence is not approval."
---

# How we built a personal data scientist for every Notion employee

[Notion](https://yomu.fyi/company/notion) · David Thomas · Oct 8, 2026

**Type:** Problem & solution

## Summary

Notion built Data Scout, an internal data agent connecting Snowflake and Notion, letting employees query governed data in plain language and cutting request turnaround from days to minutes. Within two months, weekly data team requests fell by about 80 percent, and roughly half of all employees now use it. The agent enforces Snowflake permissions per user and uses Notion business context, enabling teams such as Sales, Engineering, and Customer Experience to ask questions about account adoption, onboarding drop-off, or support volume using shared definitions. Using MCP, teams built Data Scout Reporter for scheduled reports and domain-specific agents that reuse the same context. Notion learned correct SQL could still answer the wrong question due to four failure modes, so they built governed access, curated context, instructions, query tracing, and a review process where change evidence is not approval.

## Context

Data questions at Notion required the data team’s time, with requests taking days due to triage, sourcing, and querying.

## Approach / What changed

Data Scout connects Snowflake governed data with Notion company knowledge via MCP, requiring each user’s Snowflake credentials. It provides curated definitions, business context, and a query tracing wrapper that appends metadata as JSON comments in SQL. A review process ensures changes are verified before affecting answers.

## Takeaways

- Data Scout reduced weekly data team requests by about 80% within two months and is used by roughly half of Notion employees.
- Even SQL that runs correctly can answer the wrong question due to four failure modes: wrong meaning, routing, measurement, or interpretation.
- Reliable data agents depend on governed access, curated context, and clear instructions, plus query tracing and a validation process where evidence of a change is not approval until verified by owners.

## FAQ

### How does Data Scout handle data permissions?

Each user authenticates with their own Snowflake credentials, so queries run under that person's identity. Snowflake enforces roles, grants, masking, and row-level policies, ensuring users only see data they are authorized to access.

### What failure modes can produce wrong answers despite correct SQL?

Four: wrong meaning (incorrect metric definition), wrong routing (using the wrong table), wrong measurement (ignoring historical corrections), and wrong interpretation (lacking business context). Each can cause a technically correct query to answer the wrong question.

### What workflows emerged from using Data Scout?

Two patterns emerged: recurring analysis via Data Scout Reporter, which schedules saved report definitions into Notion and Slack, and domain-specific agents providing deeper guidance for team workflows while reusing Data Scout's governed data and context.

**Tags:** [Data Governance](https://yomu.fyi/topic/data-governance), [MCP](https://yomu.fyi/topic/mcp), [Notion](https://yomu.fyi/topic/notion), [Snowflake](https://yomu.fyi/topic/snowflake)

- Source: [Notion](https://www.notion.com/en-gb/blog/how-we-built-a-personal-data-scientist-for-every-notion-employee)
- Source URL: https://www.notion.com/en-gb/blog/how-we-built-a-personal-data-scientist-for-every-notion-employee
- Ingested by Yomu: 2026-10-08T21:00:17.759Z

[Read original post](https://www.notion.com/en-gb/blog/how-we-built-a-personal-data-scientist-for-every-notion-employee)