SECTIONSPS Cloud AI Analytics Overview
PS Cloud AI Analytics enables you to leverage the power of Analytics and Einstein Discovery to analyze your professional services data. PS Cloud AI Analytics combines the data from PS Cloud Analytics and Einstein Discovery models to accurately predict project margin percentage and the number of days it takes to staff resource requests.
For more information about Einstein Discovery models, see the Salesforce Help.
Dependencies
Before you create or reconfigure a Business Analytics app to use Einstein Discovery models, you must ensure you have the following:
- The necessary licenses and permissions required to create and use datasets and dashboards with the PS Cloud Core Analytics and PS Cloud Advanced Analytics apps. For more information, see PS Cloud Core Analytics Licensing, PS Cloud Advanced Analytics Licensing, PS Cloud Core Analytics Permission Sets, and PS Cloud Advanced Analytics Permission Sets.
- The necessary CRM Analytics Plus licenses and permissions sets. For more information, see the Salesforce Help.
Limitations
Limitations on the quality of the data required to create the models are as follows:
- The value in the Project Margin (%) field cannot be the same on all project records used to create the Project Margin % Predictor (beta) model.
- The value in the Days to Staff field cannot be the same on all resource request records used to create the Days to Staff Predictor (beta) model.
Limitations on the minimum amount of data required to generate meaningful predictions are as follows:
- A minimum of 400 project records is required to generate predictions using the Project Margin % Predictor (beta) model.
- A minimum of 400 resource request records is required to generate predictions using the Days to Staff Predictor (beta) model.
Assets
PS Cloud AI Analytics creates the following assets.
|
Asset |
Name |
Description |
|---|---|---|
| Recipe | PS Cloud AI Analytics |
Pulls data from the Project Reportingdataset in PS Cloud Core Analytics. |
| Dataset | Project Margin |
Dataset created by the PS Cloud AI Analytics recipe. |
| Dataset | Days to Staff | Dataset created by the PS Cloud AI Analytics recipe. |
| Model | Project Margin % Predictor (beta) | Model created from the Project Margin dataset. |
| Model | Days to Staff Predictor (beta) | Model created from the Days to Staff dataset. |
| Dashboard | PSA Project Margin % Predictor (beta) | Dashboard that displays predicted final margin percentage for projects. |
| Dashboard | PSA Staffing Risk Predictor (beta) | Dashboard that displays predicted staffing delays for resource requests. |
Project Margin % Predictor (beta) Model
The Project Margin % Predictor (beta) model enables you to predict the Project Margin (%) field based on historical performance and determines which factors are influential in maximizing a project's margin. To do this, the PS Cloud AI Analytics recipe extracts the relevant fields from the Project Reporting dataset in PS Cloud Core Analytics to create the Project Margin dataset. Custom project fields from the Manage Additional Fields section of PS Cloud Analytics Setup are included in the dataset. This dataset is then used to create the Project Margin % Predictor (beta) model. You can then add or remove custom fields from the model and train the model to improve the accuracy of the predictions. For more information, see Training Project Margin % Predictor (beta) Models.
Once the required level of accuracy has been reached, the model can be deployed. This maps the model to the Project object and selects the most influential fields from the dataset. Predictive fields from the model are then added to the dataset, which are used to display project margin predictions for projects in the PSA Project Margin % Predictor (beta) dashboard.
Days to Staff Predictor (beta) Model
The Days to Staff Predictor (beta) model enables you to minimize the Days to Staff field from the Days to Staff dataset. Days to staff is the number of days it takes to staff resource requests on projects based on historical data in your org. The Days to Staff Predictor (beta) model analyzes the different data related to this metric in your Days to Staff dataset and highlights the factors that influence it. To do this, the PS Cloud AI Analytics recipe extracts the relevant fields from the Project Reporting dataset in PS Cloud Core Analytics to create the Days to Staff dataset. Custom project fields from the Manage Additional Fields section of PS Cloud Analytics Setup are included in the dataset. This dataset is then used to create the Days to Staff Predictor (beta) model. You can then add or remove custom fields from the model and train the model to improve the accuracy of the predictions. For more information, see Training Days to Staff Predictor (beta) Models.
Once the required level of accuracy has been reached, the model can be deployed. This maps the model to the Resource Request object and selects the most influential fields from the dataset. Predictive fields from the model are then added to the dataset, which are used to display days to staff predictions for resource requests in the PSA Staffing Risk Predictor (beta) dashboard.