Describing data modeling basics

SAP Analytics Cloud combines Business Intelligence, augmented and predictive analytics, and enterprise planning capabilities in one product so everyone has the capabilities they need to make fast, confident decisions. In order to do this, you need information from SAP Analytics Cloud stories that are based on real time or import models. You can use off-shelf-content or you can create your own but in order to so, you need modeling skills.

You have recently joined a project team that is implementing HR, sales and finance analytics using SAP Analytics Cloud. You are new to modeling in SAP Analytics Cloud and so you are interested in how the business requirements can be met and what options are available. You are taking direction from an experienced system architect on the project.

• Real time sales and HR data in stories so analytics can be up to date all of the time.
• Replicated finance data that is refreshed once per day. In other words, finance doesn’t need real time data, they only need daily snap-shots.
• • The models will use public dimensions that will be shared among multiple models.
  • Import models will be used to store the transaction data.

Stories access data from its data acquisition layer which is made up of datasets, import models and live data models.

• Project phases:
  • Phase I - Implement analytics
  • Phase II - Implement planning
• Business requirements for data access:
  • Replicated data is needed for finance
  • Real time data is needed for sales and HR

As always, the gathering of business requirements will drive the implementation process. The business requirements are normally driven by the stories that are needed for the business users. As the components are set up and the stories and applications are built, it is important to do stress testing as soon as possible particularly where the stories are complex and have high data volumes.

In addition, keep in mind how much your data will grow over time since this may affect performance.

• Data access:
  • What source systems is the data coming from?
  • Is the data accessed via live or import connections?
• For replicated data scenarios using import connections:
  • Will analytic and/or planning models be used?
  • What dimensions and measures are needed in the models?
  • What dimension properties and hierarchies are needed?
  • Will public or private dimensions be used?
  • What dimensions are needed for the sales vs. finance models?
  • Is there a need to do currency translation?
  • What are the time requirements e.g., weekly, monthly, custom hierarchies?
  • Where are calculations needed?
  • Are geo maps required?
  • What are the data access control requirements?
• For real time data scenarios using live connections:
  • Will live models will be used?
  • What are the touch points to various SAP systems?

Note

Live data can be accessed via live models or with the data analyzer without a model.

Note

Geo mapping is available for both live and replicated data scenarios, however, in this course we are focusing on the replicated use case.