---
schema_version: "1"
id: 2963d34962b413ec31539cfa5f9d6ef47cee4fedc4685f443a6697197a810a42
title: Configure Automated Prediction Writeback for a Prediction Definition
collection: admin
version: current
locale: en-us
portal_url: https://docs.salesforce.com/admin/current/en-us/analytics/264-0-0/bi_edd_model_manager_prediction_field.md
canonical_url: https://help.salesforce.com/s/articleView?id=analytics.bi_edd_model_manager_prediction_field.htm&release=264.0.0&type=5
source_site: help.salesforce.com
content_type: documentation
content_hash: c63bf01bf9a8046c5ee6e1c7a37abc830e575f12894851ca1e1f79e873251d34
description: Automated prediction writeback saves numeric prediction scores to a field in a Salesforce object. Using this feature requires that you deploy a model that outputs a numerical prediction (regression) or score (binary classification) and for which you can map all model fields back to Salesforce fields. You can use an existing AI Prediction field or create a custom field by specifying a label.
products:
  - CRM Analytics
  - Cross Cloud
release: "264"
---
# Configure Automated Prediction Writeback for a Prediction Definition

Automated prediction writeback saves numeric prediction scores to a field in a Salesforce object. Using this feature requires that you deploy a model that outputs a numerical prediction (regression) or score (binary classification) and for which you can map all model fields back to Salesforce fields. You can use an existing AI Prediction field or create a custom field by specifying a label.

1. In Model Manager, open a prediction definition.
2. In the settings panel, click **Enable** next to Automated Prediction Writeback.
    
    ![Model Manager page, showing the Enable Automatic Refresh button](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-analytics-bi-264-0-0-production-enus/ec791473-f5b4-4f64-a656-9104f455d966/bi/images/bi/bi_edd_model_manager_settings_writeback.png)
3. Select one of these following options.
    
     
    | Option | Description |
    | --- | --- |
    | **Create a prediction field** | Have Einstein create the prediction field using the prediction field label you specify. Einstein stores prediction results in this custom prediction field. You must choose a Salesforce object. _All_ fields must be mapped from the model to corresponding fields in the Salesforce object. Field mapping is supported for any field on that object for which the prediction definition was deployed, or for any parent object. |
    | **Use an existing prediction field** | Have Einstein store prediction results in an existing prediction field. |
    | **Don't store predictions** | Don’t store the prediction value in a prediction field, or store results in a custom field that you configure later. Select this option if you don’t use automated prediction fields. This kind of prediction is run programmatically via the [Get Predictions and Improvements with Einstein Prediction Service](https://help.salesforce.com/s/articleView?id=analytics.bi_edd_prediction_service.htm&release=264.0.0&type=5 "The Einstein Prediction Service REST API enables you to programmatically run predictions, manage prediction definitions, and manage models. After deploying models with Einstein Discovery, use the Einstein Prediction Service REST API to embed your predictions into any website or application.") using REST API calls or implementations. |
    
    ![Note](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-analytics-bi-264-0-0-production-enus/ec791473-f5b4-4f64-a656-9104f455d966/images/icon_note.png)
    
    Note For additional instructions, see [Display Einstein Predictions Using Automated Prediction Writeback](https://help.salesforce.com/s/articleView?id=analytics.bi_edd_prediction_automated.htm&release=264.0.0&type=5 "When automated prediction writeback fields are configured for a prediction definition that outputs a numerical prediction (regression) or score (binary classification), Einstein Discovery can save predictions directly to Salesforce records for use in reports and list views. Configure automated prediction writeback in Model Manager. After deployment, Einstein scores newly created or updated records in the Salesforce object and writes the score to the prediction field. Only the prediction is written, not the reasons for the prediction or the suggested ways in which to improve the prediction.").
4. Click **Save**. To configure performance alerts, click **Save and Next**, then proceed to [Configure Model Quality Notifications](https://help.salesforce.com/s/articleView?id=analytics.bi_edd_model_manager_notify.htm&release=264.0.0&type=5 "Models drift when characteristics in the real world data diverge significantly from the training data used to build them. Einstein can send out alert notifications if your model performance falls below configured thresholds. Significant increases in prediction error rates, missing values, or out-of-bounds values can trigger notifications, as well as detected disparate impact among variables being analyzed for bias. You can set alert thresholds, schedule weekly or monthly performance evaluation cycles, and specify who will receive notifications.").