Conversion Metric block

Use a conversion block to segment a dataset based on the observed behavior and other properties. The behavior of a dataset's record is determined by the events that are linked with the entity. The output of the conversion block is a boolean value.

The following conversion block's fields are configurable:

• Filter or Property Filter – Filters the dataset records by their property values. The left-hand side of the filter can reference a dataset property, such as a column or derived column, another block, or a parameter. Similarly, the right-hand side can reference any of these or a constant value. Comparing properties may include catalog properties such as columns, derived columns, blocks, values, or parameters. Comparison operators used in each condition are as follows.
  When working with string data, comparison semantics are case-insensitive.
  • is – The left-hand side of the condition equals the right-hand side.
  • is not – The left-hand side of the condition does not equal the right-hand side of the condition.
  • contains – The left-hand side of the condition is a string that contains the right-hand side of the condition as a substring.
  • in – The left-hand side of the condition is present within the set of values selected on the right-hand side of the condition.
  • < > >=... – Mathematical operators.
• Add single behavior – Filter the dataset to extract records associated with specific events. Use a Condition to define the criteria for filtering and an Aggregate filter to select events that meet the specified criteria.
  • Condition section definition – Select dataset records based on the presence or absence of a specific event. For instance, users can filter records for those who performed or did not perform a checkout event. The following filters are available:
    • Event Selector – Used to select specific events or broad event categories, such as an event stream.
    • Event Filter – Used to narrow down to a specific set of events.
    • Time Filter – Provides a rich interface to filter events based on their occurrence time. This filter has three modes.
      • Relative - Time is relative to the current time, for example, the last three months or all time.
      • Absolute - The time window is based on absolute dates and, optionally, time of day.
      • Relative to Event - Filters events based on the duration between an event and a baseline event or timestamp specified here. For example, within two weeks after the first occurrence of the "onboarding" event.
  • Definition of Aggregate Filter (Where) – How to limit the dataset records selection to only those where the aggregate of all matching events meets a specific condition. The aggregation process applies to the designated property of all matching events. To combine multiple aggregate filters, use the AND operator.
• Sequence – Select a dataset record if it has matching events in the specified order and meeting any associated filter criteria. Each stage of the sequence is defined by an event selection, an optional filter, and a time filter.
  • Event Selector – Used for selecting specific events or broad event categories (as an event stream).
  • (Optional) Event Filter – Used for filtering down to a specific set of events.
  • Time Filter – A rich interface for filtering events based on the time of occurrence of each event. This filter can be used in the following modes:
    • Relative – Relative to current time, for exmaple, last 3 months, or all time.
    • Absolute – Time window based on absolute dates and optionally time of day.
    • Relative to Event – Relative duration between an event and a baseline event or timestamp indiciated here. In the case of a sequence, the baseline event may also refer to a previous stage, for example, within 2 weeks after previous stage.
• Group – Any composition of previous criteria or other groups.

See Behavioral block in the Netspring documentation.