When you build segments using CRM data or custom data, you are often working with related records. This means a single contact can have multiple rows of data, such as several opportunities, purchases, or activity records.
This article explains how segmentation behaves in these scenarios, and how to structure your criteria so your results are accurate.
How segmentation works with related data
When you segment on related data, Act-On does not evaluate all conditions against a single combined record. Instead, it checks each related record individually and determines whether the contact qualifies based on any matches.
This is important because a contact may qualify even if no single record meets all your conditions.
Example: why results can be unexpected
Imagine Jon Smith has two opportunities:
| Opportunity | Stage | Amount |
|---|---|---|
| Deal A | Closed-Won | $5,000 |
| Deal B | Open | $10,000 |
If your segment includes:
- Stage = Closed-Won
- Amount > $10,000
You might expect Jon not to qualify. However, without additional configuration, Act-On can evaluate:
- Stage from Deal A
- Amount from Deal B
Because both conditions are met somewhere across the related records, Jon would still qualify.
This behavior is expected, but it often does not match how users intend to define their audience.
Use "Add Detail" to control how conditions are evaluated
Add Detail allows you to group conditions so they must be true for the same related record.
When you use Add Detail, Act-On evaluates all linked conditions together within a single row of related data.
In the example above, using Add Detail ensures that:
- The opportunity must be Closed-Won
- And the amount must be greater than $10,000
- On the same opportunity
As a result, Jon would no longer qualify.
Click Add Detail to create linked criteria within the same row.
In some cases, you may see a related table with fields that can also be linked. For example, with CRM opportunities you may also have Opportunity Product, which includes line item data related to that opportunity.
Don't see the joined tables you are expecting? Contact Act-On Support to enable them in your CRM sync.
When you should use Add Detail
Use Add Detail whenever your logic depends on multiple fields from the same related record.
Common scenarios include:
- Opportunity Stage and Opportunity Amount
- Purchase value and purchase date
- Event attendance and event type from the same row
If you are only filtering on a single field, Add Detail is not required.
Working with parent and child segments
Parent and child segments do not share record-level relationships.
This means that even if a parent segment filters on one condition, a child segment may evaluate its own conditions against different related records.
For example:
- Parent: contacts with open opportunities
- Child: contacts with opportunities over $10,000
These conditions may match different opportunities for the same contact.
To ensure accuracy:
- Repeat the full set of conditions in the child segment
- Use Add Detail again to keep them tied to the same record
Tip: If your segment depends on multiple conditions from the same object, keep them together in one row using Add Detail instead of splitting them across segment levels.
Understanding counting options for related data
When a contact has multiple related records, you also need to decide how those records are counted.
Each counting option changes how Act-On evaluates the relationship between contacts and their data.
Count all rows (no additional conditions)
This counts how many related records exist for each contact.
Example: contacts with at least one opportunity, regardless of type.
If your data includes timestamps, this may default to evaluating across all available data.
Count with criteria
This allows you to count only records that meet specific conditions.
Example: contacts with at least one opportunity where Stage = Open.
If you need multiple conditions applied to the same record, use Add Detail within the count.
None
This finds contacts with no related records at all.
Example: contacts with no purchase history.
None Where
This finds contacts that do have related records, but none match your condition.
Example: contacts with opportunities, but none are Closed Lost.
If you also want to include contacts with no related records, add an OR condition.