Connectors allow you to sync data from external sources into the Knowledge Graph automatically. Instead of manually uploading spreadsheets, you can set up a connector to pull data from an external system, map it to Knowledge Graph fields, and keep your entities up to date on a recurring schedule.

How Connectors Work

Connectors use the process of an ETL to create and update entities in the Knowledge Graph. This stands for extract, transform, load. ETL is a computing process used by many different systems. Here’s a general overview of how it works:

• First, data is extracted from a source (such as an app, a spreadsheet, an API, etc.).
• Then it is transformed into the appropriate format.
• Finally, the data is loaded into its destination. You can build connectors that load data into the Knowledge Graph or into another API. Most connectors will load data into the Knowledge Graph as entities.

Build a Connector

To build and manage connectors in Yext, go to Knowledge Graph > Connectors.

The general process for all connector types is:

1. Choose a data source. The first step in building a connector is choosing a data source to extract data from. Depending on the source you choose, you’ll have different steps to configure the data you want to pull in.
2. Add selectors. Selectors identify the parts of your data source that you want to pull in and how the data should be parsed by your destination.
   • For example, let’s say you’re building a connector to pull in HTML data from a website. You’ll use selectors to identify which parts of each webpage should be used to create or update entity data. You could specify selectors to pull in the title of the webpage as the name of the entity, map the URL of the webpage to the Website entity field, etc.
   • Each data source has a set of default selectors that you can add automatically. You can also write your own custom selectors if you choose.
3. Add transforms. Transforms can apply formatting to your data like fixing capitalization, splitting columns, standardizing phone numbers, and other changes.
4. Choose a destination. You can choose to send your data to the Knowledge Graph (as entities) or to send it to another API. Most connectors will send data to the Knowledge Graph.
5. Map fields. If you chose the Knowledge Graph as your destination, you’ll map the columns in your data set to entity fields in the Knowledge Graph. 
6. Save and run. After your fields are mapped, you’ll save your connector. You can decide to Save only, Save and Run it immediately. You can run at any time if you choose not to run the connector immediately. When you run your connector, you can configure different options like:
   • The run schedule (how frequently this connector should run)
   • The run mode (how it should handle creating new entities, and if it should delete existing entities)
   • If you want to do a dry run (to test how it would update your data without actually making changes)
     

Run a Connector

1. Navigate to Knowledge Graph > Connectors. 
2. From this screen, you can either:
   1. Select the dropdown next to the name of the connetor you want to run. Then select Run (Default Mode).
   2. Select the name of the connector you want to run. Click Run Connector, select the run mode, and click Run Connector. 

Test a Connector with Dry Run Mode

Dry runs let you run a connector and preview the results without modifying any entity data. This is useful for testing new connectors or verifying changes to an existing connector's configuration.

1. When running a connector, select your desired run mode and check the box next to Make this a Dry Run.
2. Click Run Connector.
3. After the dry run completes, it appears in the Activity Log. Click View Details to review the results.
4. Review the results in the Entity Results (Dry Run), Entity Errors (Dry Run), and Run Errors tabs.

View Connector Activity

To monitor connector runs and troubleshoot issues:

1. Navigate to Knowledge Graph > Connectors and click View next to the connector.
2. Review the Activity Log for recent runs, including status, entity counts, and error summaries.

Data Sources in Connectors

There are three main categories of data sources in Connectors:

• Connector apps
• Native sources
• Generic sources

Connector apps

Connector apps in the Yext App Directory are pre-built connectors that you can install directly in your Yext account. You can choose to customize them further after they’re installed, or use them out of the box.

Native sources

Native sources connect directly to third-party platforms to pull in data, but unlike connector apps, you still need to build the actual connector yourself.

Each native source will allow you to extract specific types of data, but they still allow for some customization.

Generic sources

Generic sources allow you to build a connector to retrieve data from a source without connecting to a specific third-party platform or app. They can take more technical work to configure, but can allow for more customization.