Maximize Customer Revenue Model Template

Analyze which types of accounts are leading to higher sales based on the last 365 days. Use a linear regression model to predict which accounts could bring additional revenue.

Note Einstein Discovery stories are now models. We wish we could snap our fingers to update the name everywhere, but you can expect to see the previous name in a few places until we replace it.

Use Case

Numerical Use Cases

Prerequisites

Einstein Discovery checks whether your Salesforce org meets the requirements to generate an app from this template. It finds your most recently closed opportunity and uses data from the year before that to generate the model training dataset.

Checks	What’s Required
Minimum requirements	At least 400 unique accounts with closed won opportunities within the first 6 months of the previous year.
Optional objects	Opportunity Line Items Leads
Object and field access	The integration user and the logged-in user have access to the objects and fields required to build an app from the template. Account Opportunity Amount Closed Date

Checks

What’s Required

Minimum requirements

At least 400 unique accounts with closed won opportunities within the first 6 months of the previous year.

Optional objects

Opportunity Line Items
Leads

Object and field access

The integration user and the logged-in user have access to the objects and fields required to build an app from the template.

Account
Opportunity
- Amount
- Closed Date

Generated Assets

Einstein Discovery generates an app containing these assets.

Asset	Default Name	Description
Model	Maximize Customer Revenue	Explains the driving factors for maximizing the customer revenue for an account, as well as recommendations on how to improve the outcome. The initial model focuses on fields that could provide good predictive signals and avoids obvious data leakage.
CRM Analytics Dataset	Account Training Dataset	Dataset used to train the machine learning model.
CRM Analytics Data Prep Recipe	`appName`_recipe	Data Prep recipe that pulls relevant account information from Salesforce to generate a dataset that is used to train a machine learning model. Design is based on best practice feature engineering and data filters.
Dashboard	About This Story	Model description Dataset description Number of Accounts Analyzed Average Customer Value Fields Created for the Training Dataset
Prediction Definition, Model	Predicted Customer Revenue, Predicted Revenue	Deployed automatically.

Salesforce Objects

Account
Opportunity
Product
Leads

Model Outcome Variable and Derived Variables

The generated dataset includes the following data. For a complete list, refer to the dataset or model.

Variable Name	Data Type	Description
`Total Revenue` (outcome)	numeric	Sum of all closed won opportunities for each account within the past three years.
`# Leads Converted`	numeric	Number of leads converted for each Account in the past number of days.
`# Won Deals`	numeric	Total number of deals won per Account based on the time frame used to calculate total revenue.
`Aggregates.FirstDealDate`	date	Close date for the first closed won Opportunity for each account based on the time frame used to calculate total revenue.
`AvgDealAge`	numeric	Average time to close a deal per Account based on the time frame used to calculate total revenue.
`AvgDealSize`	numeric	Average deal size per Account based on the time frame used to calculate total revenue.
`Top Product Family`	text	Top product family associated with each Opportunity based on the Opportunity Line Item with the largest total price.
`Top Product Name`	text	Top product associated with each Opportunity based on the Opportunity Line Item with the largest total price.