What is a data model?

Data Modeling in Sigma

Ben Harris

Technical Instructional Designer at Sigma

Welcome!

 

Welcome!

• Instructor: Ben Harris – Instructional Designer at Sigma

• Collaborators: Joshua Bowman, Andrea Cato
• Goal: Get you started with data models in Sigma

Oakmark needs your help

 

Oakmark needs your help

• create standardized models
• Avoid repeating manual work

Defining 'data models'

Defining 'data models'

 

 

• Data model = type of document in Sigma (like a workbook)
• Not focused on user interaction
• Can share data elements with other Sigma documents

Defining 'data models'

 

 

• Curating data across the organization
• Governing who sees what, consistently

Defining 'data models'

 

• Warehouse = kitchen (raw ingredients)
• Data team = staff preparing and delivering

Defining 'data models'

 

 

• Data model = the menu
• Organizes ingredients into ready-to-serve "dishes"
• End-users "order" without cooking themselves

Defining 'data models'

• Same trusted menu for everyone
• Data is accurate, governed, and reusable

Defining 'data models'

 

 

• Multiple data models in one organization
• Built by use case → team, workbooks, apps, analyses

Components of a data model

 

Components of a data model

 

 

• Start with a base table
• One row per object or event
• Example:
  • transactions → one row = one transaction

Components of a data model

 

• Base table can use multiple warehouse tables
• Join or relationships for account data
• Users add columns from other tables when needed

 

Components of a data model

 

 

• Metrics = pre-defined calculations
• Calculated values ready for workbooks

Components of a data model

 

 

• Parameters to filter data
• Column-level security
  • → data by permission

Components of a data model

 

• Advanced features (e.g. materialization) → not in this course
• By the end: firm understanding; explore more in product and docs

Let's practice!

Data Modeling in Sigma