DiscoveryGoal

Represents the metadata associated with an Einstein Discovery prediction definition.

A prediction definition is a container object in Einstein Discovery that is associated with one or more deployed models. If a prediction definition contains multiple models, then each model produces predictions for a different segment of the data. A prediction definition can contain up to ten active models. In Package Manager, this type is listed as “Discovery Prediction”.

This type extends Metadata and inherits its fields.

A DiscoveryGoal is stored in the discovery folder. DiscoveryGoals have a .goal file suffix. Here is a sample package.xml file:

1<?xml version="1.0" encoding="UTF-8"?>
2<Package xmlns="http://soap.sforce.com/2006/04/metadata">
3    <types>
4        <members>employees_Tenure</members>
5        <name>DiscoveryGoal</name>
6    </types>
7    <version>53.0</version>
8</Package>

DiscoveryGoals are available in API version 51.0 and later.

• Type: boolean
• Required: Yes
• Description: Indicates whether the prediction definition is active (True) or not (False).

• Type: DiscoveryDeployedModel[]
• Required: No
• Description: One or more deployed models associated with this prediction definition.

• Type: string
• Required: Yes
• Description: Name of the prediction definition.

• Type: DiscoveryModelCard[]
• Required: No
• Description: Model card for this prediction definition.

• Type: DiscoveryGoalOutcome
• Required: Yes
• Description: Outcome variable of this prediction definition.

• Type: DiscoveryPredictionType (enumeration of type string)
• Required: Yes
• Description: Type of prediction: Regression, Classification, or Unknown.

• Type: string
• Required: No
• Description: Automated writeback field for predictions. A custom field on the Salesforce object specified in subscribedEntity.

• Type: DiscoveryPushbackType (enumeration of type string)
• Required: No
• Description: Type of writeback field for predictions.

• Type: string
• Required: No
• Description: Salesforce object associated with this model.

• Type: DiscoveryFilter[]
• Required: No
• Description: If specified, one or more filter expressions that define the conditions under which an observation has attained its terminal state (the actual outcome has been reached). For performance monitoring, Einstein Discovery determines model accuracy by comparing a model’s predicted outcomes with actual (observed) outcomes.

Represents a field definition in custom improvement text.

• Type: DiscoveryFilter[]
• Required: No
• Description: Represents one or more filters associated with custom improvement text.

• Type: string
• Required: No
• Description: If specified, represents the user-provided template from which the custom text is computed. If not specified, then the default text is used.

Represents a model deployed in Salesforce.

• Type: boolean
• Required: Yes
• Description: Indicates whether the deployed model is active (True) or inactive (False).

• Type: string
• Required: Yes
• Description: Full name of the DiscoveryAIModel being deployed.

• Type: double
• Required: No
• Description: Threshold value. Applies only to binary classification models. For regression models, this is null.

• Type: DiscoveryFieldMap[]
• Required: No
• Description: One or more mappings between model variables and either fields (in Salesforce objects) or columns (in CRM Analytics datasets).

• Type: DiscoveryFilter[]
• Required: No
• Description: If specified, one or more segmentation filters for the deployed model. When making a prediction, the first model that has filters matching a specific input row will be used to make the prediction. No filters indicates that the model matches all input rows.

• Type: string
• Required: Yes
• Description: Label for the deployed model. Appears in Model Manager.

• Type: string
• Required: Yes
• Description: Name of the deployed model.

• Type: DiscoveryPrescribableField[]
• Required: No
• Description: Actionable fields associated with improvements.

Represents a mapping between model variables and field values.

• Type: string
• Required: Yes
• Description: Field in a Salesforce object or column in a CRM Analytics dataset.

• Type: string
• Required: Yes
• Description: Model variable.

• Type: string
• Required: No
• Description: Join key for a Salesforce object. Null if sourceType is AnalyticsDatasetField.

• Type: string
• Required: No
• Description: If the mapping is to a CRM Analytics dataset, this is the name of the dataset. Otherwise, null.

• Type: string
• Required: No
• Description: If the mapping is to a CRM Analytics dataset, this is the lookup column on that dataset used to perform the join. Otherwise, null.

• Type: DiscoveryFieldMapSourceType (enumeration of type string)
• Required: Yes
• Description: Data source type for field mapping.

Represents the data source type for field mapping: SalesforceField or AnalyticsDatasetField.

Represents a field filter.

• Type: string
• Required: Yes
• Description: Name of the field to filter.

• Type: DiscoveryFilterOperator (enumeration of type string)
• Required: Yes
• Description: Operator used to calculate the filter.

• Type: DiscoveryFilterFieldType (enumeration of type string)
• Required: No
• Description: Type of filter value.

• Type: DiscoveryFilterValue[]
• Required: No
• Description: One or more values selected for the filter.

Represents the data type of the filter field.

Represents a filter operator.

Represents a filter value.

• Type: DiscoveryFilterValueType (enumeration of type string)
• Required: Yes
• Description: Type of filter value.

• Type: string
• Required: Yes
• Description: Value.

Represents the type of filter value.

Represents the outcome variable of the model.

• Type: string
• Required: Yes
• Description: Name of the outcome variable.

• Type: string
• Required: Yes
• Description: Label for the outcome variable.

• Type: DiscoveryOutcomeGoal (enumeration of type string)
• Required: Yes
• Description: Goal for the outcome variable.

• Type: string
• Required: No
• Description: Mapped field.

Represents a model card associated with an Einstein Discovery prediction definition.

• Type: string
• Required: No
• Description: Contact email for this model card.

• Type: string
• Required: No
• Description: Contact name for this model card.

• Type: string
• Required: No
• Description: Title for this model card.

• Type: string
• Required: No
• Description: Sections in the model card.

Represents the goal for an outcome.

Represents the prediction type for a model.

Represents custom improvement text.

• Type: DiscoveryCustomPrescribableFieldDefinition[]
• Required: No
• Description: One or more strings for custom improvement text. Uses the default improvement text if none are specified.

• Type: string
• Required: Yes
• Description: Name of the model field that is actionable.

Represents the type of writeback field. Must be set to AiRecordInsight.

Here is a sample DiscoveryGoal:

1<?xml version="1.0" encoding="UTF-8"?>
2<DiscoveryGoal xmlns="http://soap.sforce.com/2006/04/metadata" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
3   <active>true</active>
4   <deployedModels>
5      <active>true</active>
6      <aiModel>Maximize_Tenure</aiModel>
7      <fieldMappings>
8         <mappedField>Opportunity.Amount</mappedField>
9         <modelField>PTO</modelField>
10         <sourceType>SalesforceField</sourceType>
11      </fieldMappings>
12      <fieldMappings>
13         <mappedField>Opportunity.ExpectedRevenue</mappedField>
14         <modelField>Salary</modelField>
15         <sourceType>SalesforceField</sourceType>
16      </fieldMappings>
17      <fieldMappings>
18         <mappedField>Level</mappedField>
19         <modelField>Level</modelField>
20         <sobjectFieldJoinKey>Opportunity.Name</sobjectFieldJoinKey>
21         <source>employees</source>
22         <sourceFieldJoinKey>Name</sourceFieldJoinKey>
23         <sourceType>AnalyticsDatasetField</sourceType>
24      </fieldMappings>
25      <fieldMappings>
26         <mappedField>Opportunity.StageName</mappedField>
27         <modelField>Field</modelField>
28         <sourceType>SalesforceField</sourceType>
29      </fieldMappings>
30      <filters>
31         <field>Opportunity.StageName</field>
32         <operator>Equal</operator>
33         <values>
34            <type>Constant</type>
35            <value>Qualification</value>
36         </values>
37      </filters>
38      <label>employees</label>
39      <name>employees</name>
40      <prescribableFields>
41         <customDefinitions>
42           <filters>
43              <field>Salary</field>
44              <operator>LessThan</operator>
45              <type>Number</type>
46              <values>
47                 <type>PlaceHolder</type>
48                 <value>[value_low]</value>
49              </values>
50           </filters>
51           <template>Increase [field_name] by [diff]</template>
52         </customDefinitions>
53         <customDefinitions>
54            <filters>
55               <field>Salary</field>
56               <operator>GreaterThan</operator>
57               <type>Number</type>
58               <values>
59                   <type>PlaceHolder</type>
60                   <value>[value_low]</value>
61                </values>
62             </filters>
63             <template xsi:nil="true"/>
64          </customDefinitions>
65          <name>Salary</name>
66      </prescribableFields>
67      <prescribableFields>
68          <customDefinitions>
69             <filters>
70                <field>Level</field>
71                <operator>LessThan</operator>
72                <type>Number</type>
73                <values>
74                   <type>PlaceHolder</type>
75                   <value>[value_low]</value>
76                </values>
77            </filters>
78            <template xsi:nil="true"/>
79          </customDefinitions>
80          <customDefinitions>
81             <filters>
82                <field>Level</field>
83                <operator>GreaterThan</operator>
84                <type>Number</type>
85                <values>
86                   <type>PlaceHolder</type>
87                   <value>[value_low]</value>
88                 </values>
89              </filters>
90              <template xsi:nil="true"/>
91          </customDefinitions>
92          <name>Level</name>
93      </prescribableFields>
94      <prescribableFields>
95         <name>Field</name>
96      </prescribableFields>
97   </deployedModels>
98   <label>employees_Tenure</label>
99   <outcome>
100      <field>Tenure</field>
101      <fieldLabel>Tenure</fieldLabel>
102      <goal>Maximize</goal>
103      <mappedField>Opportunity.Amount</mappedField>
104   </outcome>
105   <predictionType>Regression</predictionType>
106   <pushbackField>My_Pushback_Field__c</pushbackField>
107   <subscribedEntity>Opportunity</subscribedEntity>
108   <terminalStateFilters>
109      <field>Opportunity.Amount</field>
110      <operator>GreaterThan</operator>
111      <values>
112         <type>Constant</type>
113         <value>5</value>
114      </values>
115   </terminalStateFilters>
116   <terminalStateFilters>
117      <field>Opportunity.Amount</field>
118      <operator>LessThan</operator>
119      <values>
120         <type>Constant</type>
121         <value>10</value>
122      </values>
123   </terminalStateFilters>
124</DiscoveryGoal>

This metadata type supports the wildcard character * (asterisk) in the package.xml manifest file. For information about using the manifest file, see Deploying and Retrieving Metadata with the Zip File.