Reseed Documentation

Reseed is a Snowflake Native App that lets you upload and validate spreadsheets directly to Snowflake—without pipelines, infrastructure, or expanding your security perimeter.

What is Reseed?

Reseed is a Snowflake Native App that enables business users and data teams to upload spreadsheets (Excel, Google Sheets, CSV) directly to Snowflake with built-in validation and governance.

Because Reseed runs entirely inside Snowflake, your data never leaves your secure Snowflake perimeter. There's no external infrastructure to manage, no cloud buckets to configure, and no pipelines to maintain.

Key Features

• Instant validation — Missing columns, data type mismatches, constraint failures, and naming violations are caught immediately
• Reusable file patterns — Define and document the rules governing file and data ingestion across your organization using a user-friendly interface (these patterns are called Seeds)
• Query-ready outputs — Validated files are stored in a dedicated storage database as compressed CSV
• In-app access control — Build rich role hierarchies without modifying your Snowflake account's role structure
• Full customer ownership — You own the storage database; Reseed provisions it inside your account

Key Concepts

Seeds

A Seed is a data source definition in Reseed. It defines the expected structure of incoming files—columns, data types, validation rules—and who is allowed to submit or view them.

Seeds can also include naming conventions, sample data for clarity, and auto-generated data dictionaries.

Think of a Seed as a bouncer at the door.

It's alert, fair, and helpful—but firm. Every spreadsheet is checked before it gets in. If something doesn't meet the rules, it's stopped at the door with a clear explanation of why. If it does, it's welcomed inside knowing the environment is safe and orderly.

Seeds ensure that everything inside Reseed is trustworthy—so downstream systems never have to wonder what slipped through.

Uploads

When a user uploads a file to a Seed, Reseed validates the file against the Seed's rules. If validation passes, the file is stored as a compressed CSV in Reseed's storage database and optionally loaded into a table.

Roles

Reseed uses a two-tier role system: two Snowflake application roles (app_admin and app_user) control access to the app, while flexible in-app roles let you model your organization's structure without bloating your Snowflake account.

Learn more about Roles & Permissions →

Validation Rules

Each column in a Seed can have validation rules that check data quality at upload time. Rules can enforce data types, formats, ranges, patterns, and more. Invalid uploads are rejected with detailed error reports.

How It Works

1. Install Reseed from the Snowflake Marketplace

   Find Reseed in the Snowflake Marketplace and install it in your account. No external setup required.

2. Grant CREATE DATABASE

   Grant the Reseed application the global CREATE DATABASE privilege so it can provision a dedicated storage database inside your account.

3. Create a Seed

   Define the structure of the data you want to collect—columns, data types, and validation rules. You can auto-generate this from a sample file.

4. Assign Permissions

   Use Reseed's role system to control who can upload to each Seed. Users only see the Seeds they have access to.

5. Upload Files

   Users upload Excel, Google Sheets exports, or CSV files through Reseed's interface. Files are validated instantly.

6. Use Your Data

   Uploaded data is stored in Reseed's storage database. Query it directly, use it in downstream processes, or load it into tables for dashboards and analytics.

Use Cases

Reseed is designed for scenarios where you need to bring spreadsheet data into Snowflake reliably:

• Data Gaps — Capture "Mike's Spreadsheet" or "The Sheryl File" that doesn't come from your enterprise systems
• Budget Collection — Collect budget requests from departments with validation and audit trails
• Reference Data — Maintain lookup tables, mappings, and configuration data
• Pipeline Fallback — Manual upload option when automated pipelines fail
• Row-Level Security — Manage user-to-attribute mappings for Snowflake security policies
• Data Mesh — Empower domain teams to own their data without sacrificing governance

See all use cases →