This page applies to data stewards. It describes the different custom attributes that can be created in a process.

Overview

In process insights, you use processes to discover actionable insights about your organization's workflows. However, you may find that your process data doesn't have all the information you need.

When adding or editing a process, you can enhance your existing case and event data using custom attributes. Custom attributes are categorical attributes that will be analyzed along with the other attributes in your process.

Custom attributes are created on processes in process insights, so you can quickly adapt your process data without waiting for a developer to create custom record fields or make other changes to the record types used in your processes.

When creating custom attributes, you can use fields from record types and related record types as long as you have data steward access to those record types.

You can create custom attributes using any of the templates described below.

Groups Based on Aggregations

The Groups Based on Aggregations template allows you to aggregate and filter a record field from a one-to-many relationship and organize the aggregations into groups.

This template is useful when you want to categorize data from one-to-many relationships. For example, you may want an attribute that counts the number of documents in an insurance claim, then categorizes that claim as low, medium, or high complexity based on the number of documents.

See an example using this template to leverage data from a one-to-many relationship in a process.

Filtering aggregated data

When you apply a filter in this template, there will be a one-to-many relationship in the path between where you are aggregating data and where you are filtering. Because of this relationship, you will observe “at least one” filter behavior.

When you filter by a related record field from a one-to-many relationship (i.e., the "many" side of the relationship), the filter will return all records from the base record type (the "one" side of the relationship) that have at least one related record that meets the filter condition.

For example, let's say you have the Order record type and the Order Item record type, and you want to know the number of orders that have at least one order item purchased with a promo code.

Since there is a one-to-many relationship between Order and Order Item (one order can have many items purchased with or without a promo code), when you apply a filter for items purchased with a promo code, the filter will include an order as long as at least one item in the order was purchased with a promo code.

Choosing an operator

When you add more than one filter condition, you can choose how process insights should evaluate your list of filters.

You can choose any of the following operators:

Operator	Description
OR	Return records that meet one or the other filter condition.
AND	Return records that meet at least one of the filter condition.
AND ALL	Return records that meet all of the filter conditions.

Using the AND ALL operator

When you want to filter by multiple related record fields from the same one-to-many relationship, and you want to include only records that meet all of the filter conditions, use the AND ALL operator.

For example, let's say your case record type is Customer, and Customer has a one-to-many relationship to the Support Case record type. You want to filter for all customers that have at least one active support case that was created within the last week.

When you use the AND ALL operator for these filter conditions, it will only include customers with at least one related case that meets both filter conditions.

If you use the AND operator instead, the filter would still include customers that have at least one case that is both active and created in the last week. However, it would also return any customers that have at least one active case and at least one other case that was created in the last week.

Using the AND operator

When you want to filter by multiple related record fields from the same one-to-many relationship, and a record can only meet one of the filter conditions, use the AND operator.

For example, let's say your case record type is Customer, and Customer has a one-to-many relationship to the Support Case record type. You want to filter for all customers who have at least one open support case and in progress support case.

When you use the AND operator for these filter conditions, it will include customers with support cases that are open and customers with support cases that are in progress. It will not include customers with no support cases or customers with support cases in other statuses.

If you use the AND ALL operator instead, the filter would not include any customers. Because a support case cannot be both open and in progress, there will not be any records that meet all of the filter conditions.

Groups Based on a Range

The Groups Based on a Range template uses a range to organize your numerical data into groups. This template only supports record fields of type Number (Integer) or Number (Decimal).

This template is useful when you want to categorize data based on a range of values. This can also be used to hide specifics of sensitive data. For example, you can create a custom attribute that categorizes employee salary ranges rather than displaying salaries as exact values.

See an example using this template to hide sensitive data from a process.

Groups Based on a Date Difference

The Groups Based on a Date Difference template allows you to calculate the difference between two dates and organize the difference into groups. This template requires at least one field of type Date or Date and Time.

This template is useful when you want to categorize how long it took to complete something in your business process. For example, you can create a custom attribute that categorizes deliveries by whether they're on time or overdue.

You can also categorize events relative to a specific date. For example, you can create a custom attribute that calculates the difference between a payment date and a due date, then categorizes the payment as early, on time, or past due.

See an example using this template to clean up process data.