Einstein Analytics Single Page Overview

The objective of this post is to put together the overall architecture and basic building blocks of Einstein Analytics. This post excludes Einstein Discovery Insights.

Einstein Analytics The Big Picture

\"Einstein

Basic Building Blocks

  1. Data Sources: Source Systems of your data that we will be analyzing using EA
    • Local Salesforce Environment:
    • External Salesforce Environment
    • External System(s)
    • CSV extracts
  2. Connection: Connection becomes available once Data Sync is enabled. Without Connection, only the Local Salesforce Environment is available as a data source. Each connection represents one source system. Even if the local Salesforce environment is the only data source, enabling Data Sync decouples data extraction and transformation and improves performance.
  3. Data
    • Connected Data: Each of the data tables/objects is represented as one connected data object. Connected data objects are not available directly in a lense or dashboard, it must be transformed using a data flow or a recipe first.
    • Datasets: Datasets are the end result of transformation using a data flow or a recipe.
    • Salesforce direct: Objects from the local Salesforce org are available for use using Salesforce direct in a dashboard.
  4. Transformation
    • Dataflow: A dataflow can consume one or more Connected data and datasets, perform various transformation operation and generate one or more datasets. There are limits on the total number of data flows and execution time of long-running data flows, so it is recommended to optimize the transformations and minimize the total number of data flows.
      \"Einstein
    • Recipe: A recipe can consume one or more Connected data and datasets, perform various transformation operation and generate one dataset. Similar to the dataflows optimization of the transformation and minimization of the total number of long-running recipes is the key. One key benefit of the recipe over data flow is the easy visual while the transformation being performed as the UI is more visual with what is happening with the data and there are few transformations that are only available in Recipe.
      \"Einstein
  5. Analyze
    • Lense: explore one dataset to gain insight.
    • Dashboard: Perform advanced exploration of one or more datasets or the salesforce objects from the local salesforce org directly.
    • Application: Combine Dashboards, Lenses, and Datasets in an application and share with specific users, groups, roles, and roles & subordinates.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top