1. Determine the Appropriate Pipeline

Running a synchronization process normally involves running a pipeline batch. But pipeline batches are designed to handle sets of data, and to be run asynchronously.

In this case, the synchronization only handles a single object, and it needs to run synchronously. Specifically, it needs to handle a single xDB contact.

The pipeline batch xDB Contacts to CRM Sync Pipeline Batch contains this logic. By looking at the pipelines that this pipeline batch runs, you can see that the pipeline Upsert Single CRM Contact Pipeline is responsible for handling a single xDB contact. This is the pipeline you are looking for.

Note the item ID for the pipeline definition item.