Loading
Feature degradation | Gmail Email delivery failureRead More
CRM Analytics
Table of Contents
Select Filters

          No results
          No results
          Here are some search tips

          Check the spelling of your keywords.
          Use more general search terms.
          Select fewer filters to broaden your search.

          Search all of Salesforce Help
          Handle Live Data Alerts

          Handle Live Data Alerts

          Live data monitoring detects when your production data contains high levels of missing or out-of-bounds values—indications that your model’s accuracy is declining. Model Manager surfaces alerts so that you can easily see what’s happening in your data.

          Note
          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.

          To handle live data alerts:

          1. In Model Manager, open a prediction definition.
            Einstein Discovery shows you when detected missing or out-of-bounds values exceed configured thresholds.
            Model performance monitoring alerts
          2. To investigate, click either Review Alerts or the link in the Alerts column.

            The Live Data Alerts window shows you alerts with tabs for each model in the prediction definition.

            Data alerts window
            Bar charts show you occurrences of data issues by outcome value. Orange represents the number of rows with issues (out-of-bound or missing values).
          3. To see details for an outcome value, hover over its bar in the chart.
            Hover over a bar to see details.
          4. Address any issues.
            • To review and tune model settings, click View Model at the bottom of the screen.
            • To address out-of-bounds issues, consider refreshing the CRM Analytics dataset with data that better represents the outcome you want to predict.
            • To address missing values, consider using the Replace Missing Values transformation in model settings.
            When finished, create a model version to automatically train the model, and then deploy and update the model in Salesforce. Einstein automatically resets performance monitoring statistics for the latest model version.
           
          Loading
          Salesforce Help | Article