Understanding the Issue

The issue arises because Unified Individual Contact Point Email (and Address/Phone) has a direct relationship with the Unified Individual, but also an indirect one. This can lead to different results when using the Direct attribute versus the Related attribute in a segment.

The root cause of this issue is the way the segment builder handles the relationships between the Unified Individual and the Contact Point information. The Container Path may be valid, but the actual data may not contain the correct joining values, leading to unexpected results.

Fixing the Issue

To segment on the preferred contact point information, you can use a data transform to create a new DMO with a relationship with the Unified Individual, and select the email you wish from your criteria. Another option is to use a Calculated Insight with the number of internal email addresses.

Data Transform Example

/* Create a new DMO with a relationship with the Unified Individual */
var dmo = DataModelObject.create('PreferredEmail');
dmo.addRelationship('UnifiedIndividual', 'id');
/* Select the email you wish from your criteria */
dmo.addField('email', 'string');

Alternatively, you can use a Calculated Insight to achieve the same result. This approach can be more efficient, especially when dealing with large datasets.

Best Practices

Heads up: when segmenting on contact point information, make sure to consider the potential for bogus or invalid data, such as email addresses that are not legitimate.

Segmentation Checklist

• Use a data transform to create a new DMO with a relationship with the Unified Individual
• Select the email you wish from your criteria
• Consider using a Calculated Insight for more efficient segmentation
• Be aware of potential bogus or invalid data
• Test your segments thoroughly to ensure accurate results
• Use the preferred contact point information to ensure accurate segmentation

Frequently Asked Questions

What is the difference between Direct and Related Attributes in segmentation?

The Direct attribute refers to the direct relationship between the Unified Individual and the Contact Point information, while the Related attribute refers to the indirect relationship.

Can I use a Calculated Insight to segment on contact point information?

Yes, you can use a Calculated Insight to segment on contact point information, and it can be a more efficient approach than using a data transform.

How do I handle bogus or invalid data when segmenting on contact point information?

You should be aware of the potential for bogus or invalid data and take steps to handle it, such as using data validation or data cleansing techniques.

What is the best way to test my segments to ensure accurate results?

You should test your segments thoroughly to ensure accurate results, using a combination of data validation and testing techniques.