The timeline event rollups feature (accessible via More > Timeline events > Timeline event rollups) offers a Basic query builder that allows you to roll up data stored in timeline events and save the results to a profile property.
When creating a rollup, you first select your timeline event type and then add any conditions that your timeline events must meet.
For these conditions, the Basic query builder offers different operators to match timeline event properties to values; the options available depend on the data type of the selected property.
If your timeline event property uses the Text data type, then you can use the operators "Equals" or "Does not equal" to match that property to specific text values. The following sections provide more detail on how the Basic query builder matches text using these operators.
Equals
Use the "Equals" operator when you're looking for an exact match of your property value (i.e., the property contains any of the values you enter). It is not case sensitive, so capitalization is not taken into account.
For example, you have the following condition:
"Country" EQUALS Mexico OR Netherlands
The condition is only met if the property value is either Mexico or Netherlands, since those are exact matches. The condition would NOT be met for any other value, even for misspellings.
This is illustrated in the following table:
Property value | Is the condition met? |
Mexico | Yes |
Netherlands | Yes |
mexico | Yes |
netherlands | Yes |
[An empty value] | No |
France | No |
Frnace | No |
Mecixo | No |
Netherlans | No |
If your timeline event property contains multiple values, the rollup only uses the first value (unless the property is flagged as "multi-value," as outlined below in the section Multi-value properties). For example:
Property values | Result |
Netherlands and Mexico | Netherlands is rolled up, since it meets the condition as the first value. |
France and Netherlands | No data is rolled up, since France does not meet the condition. |
Mexico, Netherlands, and France | Mexico is rolled up. |
Does not equal
Use the "Does not equal" operator when you're looking to avoid certain exact matches (i.e., the property does not contain any of the values you enter). This operator is also case insensitive.
For example, let's use the same condition with "Does not equal":
"Country" DOES NOT EQUAL Mexico OR Netherlands
The condition is met if the property value is anything other than Mexico or Netherlands. If the value is either of those two countries (with the exact spelling), the condition would NOT be met.
This is illustrated in the following table:
Property value | Is the condition met? |
Mexico | No |
Netherlands | No |
mexico | No |
netherlands | No |
[An empty value] | Yes |
France | Yes |
Frnace | Yes |
Mecixo | Yes |
Netherlans | Yes |
As with the "Equals" operator, if your property contains multiple values, the rollup only uses the first value (unless the property is flagged as "multi-value"). For example:
Property values | Result |
Netherlands and Mexico | No data is rolled up, since the condition is not met with Netherlands as the first value. |
France and Netherlands | France is rolled up. |
Mexico, Netherlands, and France | No data is rolled up. |
Multi-value properties
If you add a multi-value property to a rollup condition, the rollup takes into account all of the property's values (not just the first one). By default, these timeline event properties are flagged as "multi-value":
Order event type > Tag property
Order event type > Product property > Product tag subproperty
Order event type > Product property > Product category subproperty
Order refund event type > Product property > Product category subproperty
Abandoned basket event type > Product property > Product category subproperty
For instance, you have the same condition as above:
"Country" EQUALS Mexico OR Netherlands
If "Country" were a multi-value property and that property contains the values France and Netherlands, the condition would now be met. Both values would be rolled up.
Use cases
This functionality is especially useful for BlueConic customers that want to better target consumers and have event types in their tenant with nested properties and subproperties. For example, if a clothing company sells a men's leather belt across three different product categories—"men's products," "belts," and "leather products"—all three product categories can now be rolled up when using the Basic query builder, since "product category" is a default multi-value property.
This rolled-up data can then be used for top retail use cases such as smarter product recommendations and better content personalization.