Add Matched Entities
A matched entity allows you to access relevant fields from a data source that's not the intake source for the module. To access these fields, you create a link between two pieces of data from different data sources. This link defines two fields as having the same value, providing a shared reference point from which other data can be retrieved and used in your workflows.
For example, imagine you have a module that monitors a customer support ticketing system like Zendesk as the intake source and you want to notify the manager of the agent assigned to the Zendesk ticket upon completion. The manager's email address isn't in Zendesk, but you can access it in the ERP system, Workday. You can therefore create a matched entity that connects Zendesk and Workday, using the piece of data that both data sources share: the agent's email address. You can connect the assignee email address field in Zendesk to the user's email address field in Workday, telling Tonkean these fields are equal.
Once the data sources are connected, you have access to the other fields in Workday, including that user's manager. You can then retrieve that manager name and email address, and use that email address in a Send Notification action. You now have access to that field can can use it anywhere in your module. Any of the fields from Workday can be used in a similar way.
At its core, entity matching is a simple concept—you're simply adding one or more fields to an item that come from a different data source—but the possibilities for using matched entities are almost endless. Matched entities are the primary method for connecting additional data sources beyond the intake source for this reason.
Common Use Cases
To better understand matched entities and what they allow you to do, it's helpful to look at some real-world use cases:
Employee's Out of Office? Go to Plan B
Context - The Risk Management team at a large hotel chain has a module that monitors a shared email inbox as its intake source. They receive various kinds of requests in this inbox, including those regarding insurance, invoices, and employee health and safety. The module uses NLP to evaluate the content of each email coming in and determines the right category for the ServiceNow ticket and the person it should be assigned to.
Problem - Since the module automates the assignment of ServiceNow tickets to various members of the Risk Management team, it can sometimes assign a ticket to an employee who's out of the office, essentially stranding the ticket and leaving the requester without any progress on their problem.
Solution - A separate module sends employees a form where they enter their out-of-office dates and their designated back-up person. The module captures this data in an Excel spreadsheet.
The first module uses this Excel spreadsheet as a matched entity using the employee's email as the common field between ServiceNow and the Excel spreadsheet. When the module is about to assign the ServiceNow ticket, the matched entity checks the Excel spreadsheet to see if the employee email and the current date matches. If a match is found, we know the employee is current out of office and the module assigns the ticket to their designated backup (found in the Excel spreadsheet) instead.
Who's emailing us? What do they need?
Context - The Marketing team at an enterprise automation company has a module that monitors an email inbox as its intake source, where they primarily receive requests for information about product offerings, demos, and other similar requests.
Problem - The senders of these emails vary quite a bit—from existing customers asking about enhancements or upgrades to brand new contacts interested in the product. Marketing wants to lead each audience down the right path to help increase sales and drive engagement, and it's too labor-intensive to check every email individually.
Solution - Create two matched entities that check the email sender against two entities from an existing data source: (1) Sale leads in Salesforce and (2) Contacts in Salesforce. If the sender is matched as an existing lead, the module initiates an email campaign, but if the sender is matched to an existing contact, the module assigns the email to a member of the Marketing team so they can respond personally.
Additionally, the module also takes into account if no match is found in either the Salesforce sales leads or contacts entities. If no match is found, the module creates a new sales lead and initiates a top-of-funnel email campaign.
Best Practices
Matched entities are an incredibly powerful feature and are an integral part of most complex workflows. Below are some best practices we recommend to avoid common mistakes and get the most out of matched entities.
When setting up a matched entity and using it in your workflow, always account for the possibility that there may be no match found. Build in logic that handles this situation, whether it be creating a new entity in the relevant system or reaching out to a person to take further steps.
When working with large data sets where there's a possibility of multiple matches in a single entity, define how the module should proceed. See step #11 in Create a Matched Entity for details on configuring this logic.