Thresholds are the rules that dictate what causes a given node to run.
Dependents are follow-up actions that occur based on a previous node in the Workflow. This is done to create a sequence of actions (set a tag, run another service, apply an outcome on the evaluation) based on certain conditions or attributes returned by a previous node.
This article will cover the following topics on thresholds, dependents, and how to update them in further detail:
- Understanding the Terminology: Thresholds and Dependents
- Adding Dependent Nodes
- Managing Threshold Rules
- Viewing a Node’s Dependents
Understanding the Terminology: Thresholds and Dependents
In a Workflow, a "threshold" refers to the rule or set of rules that determine when a particular node will execute. Some nodes may not have thresholds, meaning they always run, and appear on the left-most edge of the Workflow. However, by applying threshold rules, you can customize a node to run only under certain conditions. For example, suppose you want to create a tag called "Minor" to identify applicants under the age of 18. Without any threshold rules, every application would receive this tag, regardless of the applicant's age, making the tag inaccurate. By setting a threshold rule such as "age is less than 18," the tag will only be applied under a specific condition: to applicants who are actually minors.
Within the context of a Workflow, a "dependent" is referred to as the subsequent action that occurs from a given node. One node is dependent on another node running. This action can be: setting another tag, running a data service, contributing to a Workflow outcome, and more. Therefore, dependents are the follow-up actions or checks that occur once a particular condition is met that triggers another node to run. Dependents can have their own dependents as well.
By adding a threshold rule to the Workflow, you would also be adding dependents.
Using a simple example, consider this Workflow:
The "Denylist" service is configured to run. However, on its own, the "Denylist" service simply returns data. A follow-up action would need to be configured to trigger an action based on the data returned.
The "Denylist" service is the initiating node and has several follow-up nodes after the Denylist service run: setting the tags "State Denylist Denied" and "IP Denylist Denied". These tags are dependents of the "Denylist" service running. What actually causes the "State Denylist Denied" and "IP Denylist Denied" to get set on an application are based on the threshold rules configured for these tags.
Both of these tag nodes also have their own dependents, and they share the same dependent. The tag "Denylist Alert" is a dependent of both "State Denylist Denied" tag and "IP Denylist Denied" tag. What causes the "Denylist Alert" tag to get set on an application is based on the threshold rules configured for this tag.
Adding Dependent Nodes
Adding a New Dependent
A dependent always occurs off an existing node that’s already part of the Workflow. To add a new dependent:
-
Hover over the desired node. A plus icon will appear to the right.
Note: If another node is already in focus (clicked into a node and the node panel has slid out), the plus icon will be replaced by a link icon. To exit a node, click anywhere on the page or directly click the intended node.
-
Click the plus icon (Add Child Node) to create a dependent from the parent node.
-
A placeholder node will appear in the Workflow editor and a drawer to build out the threshold rules will slide out from the bottom. The threshold logic will be started based on which node you clicked on in the previous step.
Linking Additional Dependents to an Existing Node
Once a node has been added to the Workflow, additional threshold rules can be linked to continue layering rules in additional conditions at any time. You have as much much control as you need to define when an action occurs on a given node.
- While clicked into a node you’d like to add more rules to, hover over any other node. A link icon will appear to the right. Note: If there is no node already in focus, the link icon will be replaced by a plus icon. You must directly click the intended node you’d like to add more rules to.
- Click the link icon (Link to Node) to add a node as a new threshold rule.
A new blue line will flow between the nodes to visually show the relationship.
Managing Threshold Rules
Once a node is created, the node panel can be opened to the Threshold tab at any time by clicking on the node to manage the existing rules. There are several options within the Threshold drawer to support adding or modifying the existing conditions. This section aims to cover the different functionality within the Threshold drawer.
Add a New Row of Logic
Additional dependents can be added by following the steps under Adding Dependent Nodes. Alternatively, for convenience, a new row can be added in the desired location within the Threshold tab and adjust the rule from there.
- Click on the line-number in the numbered side bar where you’d like to insert a new row.
- In the drop-down menu, click Add Above or Add Below based on the preference of where the new row should be added.
- A new row will be added immediately in the preferred row, the default logic added is the logic in Row 1.
- From here, edit the threshold condition in the newly added row by clicking on any field highlighted in yellow on hover.
Delete Threshold Rule
Once there are at least two rows of thresholds rules, unwanted rules can be deleted.
- With two or more rules, click on the line-number for the row you’d like to remove.
- In the drop-down menu, click the Delete option.
- The row will be immediately deleted.
Reorder Placement of Rows
After adding at least one line of logic, the numbered side bar becomes available. This numbered side bar also serves as movement of a threshold logic line. If the order of the threshold rules needs to be adjusted, each row can be drag-and-dropped. This is useful when more complex logic with nesting is being built, where specific order does matter.
- Click and hold the line-number in the numbered side bar.
- Start dragging the row to the desired new placement. Dragging is in effect when you can see the row being dragged with the mouse cursor.
- As the row is dragged, all valid placements are highlighted in yellow.
- Release the mouse click over the desired placement to place the row in its new desired place.
Adding and Removing a Group
Groups allow a threshold or attribute expression to have a clearer order of logic that the Workflow should evaluate. Without Groups, complex logic with mixed AND and OR operators can be processed incorrectly or outside the intended order.
Take for example this Threshold expression:
Based on the above, the order of processing remains confusing. The complexity and confusion increases as more AND and OR operators are mixed together without Groups.
However, with Group logic added, it’s clearer what the combination of logic is intended to be.
To reduce ambiguous logic, the Workflow editor will automatically add a Group when there is a mix of AND and OR operators. For more complex logic which requires additional layers of nested grouping logic, follow these steps to create a Group.
- Click on the yellow space in the desired location where the Group should be opened.
- Click Open Group.
- If starting a Group was not intended, you can click Cancel opening this group or the Esc key on your keyboard.
- Click on the yellow space in the desired location where the Group should be closed.
Groups can be nested as many layers as desired.
To remove a Group after adding it:
- Click on an open or closed parenthesis.
- Click Ungroup this.
Once a Group has been ungrouped, the Workflow editor will automatically adjust the logic to reduce ambiguous logic. You may notice a previous OR operator is changed to an AND operator, or vice versa where a previous AND operator is changed to an OR operator depending on the logic.
Viewing a Node’s Dependents
Once additional nodes are added to the Workflow, a parent-child relationship builds between the nodes. These relationships can be seen in the Dependents tab of the node panel. The node you’re viewing can be considered the "parent" whereas the dependents are direct "children" of the node. The tab reflects which follow-up dependents in the form of tags, outcomes, data services, matrixes rely on the node being viewed to execute.
To view a node’s dependents:
- Click on the node you’d like to review.
- Once the panel opens, click the Dependents tab.
A node without any dependents relying on it will say "This action has no dependents."
Whereas a node that does have dependents will show the number of dependents it has in the tab name. Clicking on the Dependents tab will list out all actions dependent on the node.
Here are helpful features to identify each dependent is under the Dependents tab:
Node type | Distinct Identifier | Example |
Outcome: Approved | The Approved outcome will be underlined with a light green color that matches the "Approved" outcome in the Outcomes panel. | |
Outcome: Denied | The Denied outcome will be underlined with a red color that matches the "Denied" outcome in the Outcomes panel. | |
Outcome: Manual Review | The Manual Review outcome will be underlined with a yellow color that matches the "Manual Review" outcome in the Outcomes panel. | |
Outcome: custom outcome | Custom outcomes will be underlined with a grey color that matches the color of custom outcomes in the Outcomes panel. | |
Tag | A Tag name be underlined with the color chosen for the tag. Tag colors themselves are arbitrary, but a color can be used to denote the health or risk of the applicant, whether the tag is informational, or any other desired classifications. | |
Data service | Written in code format and text color is green. | |
Matrix Models | Written in code format and text color is blue. |
Any dependent can be hovered over to preview the threshold logic. Clicking on the dependent will take you directly to the node’s or outcome’s Threshold rules.
Comments
0 comments
Article is closed for comments.