Data layers in Whaly
Whaly offers you 4 main layers to expose your data to stakeholders. Each layer is the foundation for the next one.
Here is a diagram with the 4 data layers offered in Whaly ⤵️
Why 4 data layers?
Offering 4 layers is provided the proper tools to your organization to properly govern your data assets and make sure that your Business Intelligence can be maintained and remains trusted by all data consumers in the long run.
Each of the data layer is offered with a set of tools that will enable your organization to remain in control of your Business Intelligence deployment and remain effective even in the long run.
Layer 1 - Sources / Raw data
At the very base of Whaly Data layers is the "Source" layer. This is where data is being sourced from to be then exposed in the Business Intelligence. Think of this layer as being the "Import layer" of Whaly. Everything analyzed in Whaly will be ultimately linked to a Data Source.
In this layer, it'll be important to:
Import all the data sources that are needed to build business insights from
This layer is managed by Data Builders in Whaly through the Workbench interface.
Depending on the state of your Data Stack and of your current deployment, the Source layer will enable you to:
a. Import Data already existing in your Data Warehouse (that came from an ETL vendor or from a custom pipeline)
b. Import Data managed by a modelling layer such as dbt
c. Import Data directly from one of your Business Tools with an integrated connector of Whaly
Layer 2 - Models
Above the Sources layer, exist a Models layer. This layer will help you clean and consolidate your various data sources to produce Models that will be analyzed and reported on by your Data Consumer.
In this layer, it'll be important to:
Build the "Data Model" of your organization, by merging different sources and building relationships between different business entities (ex. Customers and Orders)
Document the work that has be done to implement your business rules and definition
Produce data in a shape that will be easily analyzed by your Data Consumer
This layer is managed by Data Builders in Whaly through the Workbench interface.
Depending on the state of your Data Stack and of your current deployment, the Models layer will enable you to:
a. Import models managed by a modelling layer such as dbt
b. Build SQL based models directly from Whaly interface
c. Build Flow models with a no-code interface to bootstrap your models at the beginning of your Data Projects
Layer 3 - Explorations / Semantic Layer
Above the Models layer, exists the Exploration layer. This is also called "Semantic Layer". This layer will be the place where Data Builders and Data Consumers will collaborate. It will be the interface designed by Data Builders for Data Consumers to enable them to make self service queries.
In this layer, it'll be important to:
Design the proper set of Dimensions / Metrics that will empower your whole team
Document the Self Service interface to facilitate the adoption of Data Consumers and make them autonomous
This layer is configured by Data Builders in Whaly through the Workbench interface and used by Data Consumers through a dedicated query interface
Layer 4 - Reports
The Reports layer are where your Data Consumer will build their own dashboards and questions to suits their business needs. In this layer, they will be able to manage the look & feel of their data visualization, manage the sharing methods (pushs to external systems, sharing links, invitations).
A folder structure and access control settings will help your Data Consumers organize their Dashboards in a content hierarchy to maximize the adoption of the dashboards.
In this layer, it'll be important to:
Design and shape dashboards and charts so that they expose the right information and help everyone take better decision
Structure the overall content structure of your organization to make sure that people can find the insights they are looking for
This layer is configured by Data Consumers in Whaly in the Workspace interface
Last updated
