Get Einstein Discovery predictions on records in your Lightning Reports—directly in Slack. Use the CRM Analytics for Slack app to run an Einstein Discovery model on the records in the report. The app returns the average prediction score, top-influencing factors, recommended actions, and the most at-risk records in the report. Share the predictions with your Slack channels and colleagues.
Required Editions
| User Permissions Needed | |
|---|---|
| To use CRM Analytics for Slack: |
Connect Salesforce with Slack |
| To view Einstein Discovery predictions, improvements, and related details that are derived from the deployed model: | View Einstein Discovery Recommendations |
Prerequisites
A CRM Analytics Plus or Einstein Predictions license is required to access Einstein Discovery. See License and Permission Requirements for Einstein Discovery.
In CRM Analytics, create an Einstein Discovery story and deploy the model to a Salesforce object with all the fields mapped. See Deploy Models.
In Salesforce, create a report containing the records you want to analyze. The report must contain the record ID or name. Make sure the report’s primary object matches the object where the Einstein Discovery model is deployed. See Build a Report.
Run Einstein Discovery Predictions in Slack
To view and share predictions in Slack:
- In the CRM Analytics app for Slack, find the report and select Run Prediction in the action menu.
- View results in the Predictions window. See the table for result descriptions.
- Optionally, scroll down and click Share to collaborate directly in Slack.
| Result | Description |
|---|---|
| Average Prediction Score |
The overall predicted likelihood to reach the outcome defined in the model, across the first 2000 records in the report. In Einstein Discovery, a prediction is a derived value, produced by a model, that represents a possible future outcome. You can think of a prediction as the output of a predictive model that is based on the inputs of predictor variables that the model accepts. |
| Top Predictors | The conditions that most significantly drive the predicted outcome, in decreasing order of magnitude. A condition is a data value associated with a column. In Einstein Discovery, a predictor consists of one or two conditions. |
| How to Improve this Outcome |
Suggested action, based on prescriptive analytics, that a user can take to improve the likelihood of a desired outcome. Improvements are associated with actionable variables, which are explanatory variables that people can control. Taking a suggested action can improve the predicted outcome. An improvement is analogous to a prescription in prescriptive analytics. Improvements are shown based on the model's actionable variables. Actionable variables are optional. If no actionable variables are configured, improvements aren’t shown in results. |
| Records Least Likely to Reach a Successful Outcome |
The records in the report with the strongest predicted likelihood to be unsuccessful, among the first 2000 records in the report. Success is based on the goal defined in the model. Requires binary classification model. |