← Back to knowledge base

Getting Started

Connect Data Sources (Snowflake, Power BI, Google Sheets, Files)

Setup and validation guide for all supported DashboardGenius data sources.

Connect Data Sources

Reliable analysis starts with a clean connection and a quick validation pass.

Supported Source Types

  • Snowflake for warehouse-backed operational data
  • Power BI for existing semantic models and leadership datasets
  • Google Sheets for fast-moving trackers and shared operating sheets
  • File uploads for ad hoc exports and reference material

You can connect more than one source for the same organization. That is often the best setup when one system holds trusted production facts and another holds planning, shortage, or business-impact context.

Snowflake Connection

What you enter

  • Account
  • Username
  • Password
  • Warehouse
  • Database
  • Schema

Best fit

Use Snowflake when you want the broadest direct access to warehouse-level production, quality, or operational data.

Snowflake is usually the best first connection when your team wants broad exploration across many operational tables instead of one pre-modeled business view.

RedZone customer flow

If RedZone Customer is enabled during setup:

  • Some connection fields are pre-filled
  • Database can be derived from username format
  • RedZone knowledge-base setup may run after successful connection

Best practices

  • Use a dedicated analytics user when possible
  • Confirm credentials with your internal data owner
  • Keep naming clear so connected sources are easy for your team to identify
  • Validate one known KPI before inviting broader trust in the connection
  • If you connect multiple Snowflake sources, name them by plant or business lane so follow-up questions stay clear

Power BI Connection

What you enter

Enter:

  • Tenant ID
  • Client ID
  • Client Secret
  • Workspace ID
  • Dataset ID

Best fit

Use Power BI when your team already relies on a specific dataset or semantic model for executive reviews and you want DashboardGenius to stay anchored to that source.

Power BI is usually the best fit when leadership already trusts a curated business model and wants DashboardGenius to stay aligned with that same semantic layer.

Best practices

  • Connect the same dataset your team uses in operational reviews
  • Use clear connection names by plant or function
  • If multiple datasets exist, start with the one that already answers your core weekly meeting questions
  • Keep the source description business-oriented so teammates know what kinds of questions belong there

Google Sheets Connection

Setup flow

  1. Share the sheet with the service account email shown in onboarding
  2. Paste the Google Sheet URL
  3. Connect and test

Best fit

Use Google Sheets for fast-moving trackers such as shortages, demand signals, production plans, action logs, or business-impact context that changes more often than warehouse pipelines.

DashboardGenius reads current sheet data during queries, so teams can use this for frequently updated operational trackers without waiting on a warehouse refresh.

Best practices

  • Share only the sheets relevant to the workflow
  • Use clear tab names and header rows
  • Keep codes, item names, and date columns consistent
  • Avoid mixing unrelated trackers into one workbook if different teams own them
  • If a workbook supports a specific workflow, describe that clearly in the source name or routing notes

File Upload Connection

Use Upload Files when you need rapid analysis on source files.

Notes:

  • Multiple files are supported
  • Onboarding file uploads enforce a per-file size limit shown in UI
  • Keep filenames descriptive (for example: Line-2-Downtime-2026-03.csv)
  • Use files for focused analysis, not as a replacement for a well-maintained live source

Files are especially useful for one-time exports, reference PDFs, and supporting material that helps explain a situation without becoming a permanent live source.

Choosing the Right Source First

Start with the source that already answers your most trusted KPI question.

  • Choose Snowflake when breadth and raw access matter most
  • Choose Power BI when a curated business model already exists
  • Choose Google Sheets when operational context changes quickly
  • Choose Files when you need fast one-off analysis on exports or supporting material

If you are unsure, start with the source your team already uses in a weekly operating review. Trust usually matters more than raw breadth on day one.

After You Connect: Keep the Source Understandable

A connected source is only half the setup. Good source hygiene makes DashboardGenius easier for both people and AI to use.

After connection, review:

  • Source name
  • Business description
  • Whether the source belongs to a clear domain such as operations, quality, demand, logistics, or finance
  • Whether there are obvious aliases your team uses for the same concepts

This matters most when your organization has multiple Power BI datasets, multiple Google Sheets workbooks, or a mix of warehouse and tracker data.

Sources and Routing Map

DashboardGenius includes two admin views for ongoing source management:

  • Sources to inspect connected sources, rename them, and review their summaries
  • Routing Map to document what each source is best for so future questions route more cleanly

Use Routing Map when you want to clarify:

  • what business domain a source covers
  • alternate names or shorthand people use
  • the kinds of questions the source should answer well
  • situations where a source should not be the first choice

This is especially helpful after your first few weeks of usage, once you know where people get confused.

If Schema Discovery Fails

For Snowflake and Power BI flows, you may be prompted to enter table names manually.

Use business-friendly, known table names and reconnect. Then run validation prompts before production use.

Cross-Source Analysis Tips

Once individual sources are validated, you can ask cross-source questions. Keep the request explicit so DashboardGenius knows what should come from which source.

Good example:

"Compare Snowflake downtime trends with the shortage tracker in Google Sheets for the last 14 days."

Bad example:

"Use everything and tell me what matters."

Cross-source analysis works best when each source has a distinct job. For example:

  • Snowflake for production facts
  • Power BI for curated executive KPIs
  • Google Sheets for shortage, plan, or business-impact context

Validation Checklist (Run Every Time)

After any new or edited connection:

  1. Confirm expected date range appears
  2. Confirm key tables/tabs/datasets are visible
  3. Ask one known KPI question and compare with your trusted source
  4. Ask one scoped follow-up question to confirm filters and breakouts behave as expected

If you changed the source name or routing details, ask one business-language question too. That confirms the source is understandable in the same words your team actually uses.

Naming Convention Recommendation

Use this simple format:

  • Plant - Source - Domain

Examples:

  • Plant A - Snowflake - Ops
  • Corporate - Power BI - Executive KPI
  • Line 4 - Sheets - Quality Log

Avoid generic names such as dataset 1, sheet copy, or prod export final.

Common Setup Mistakes

  • Connecting a source but never validating a known KPI
  • Giving multiple sources vague or overlapping names
  • Using one workbook for many unrelated workflows with no clear tabs or ownership
  • Expecting file uploads to replace a durable live-source setup
  • Asking broad cross-source questions before each source works well on its own

Next Guide

Continue with Manage Data Sources and Routing Map.