> ## Documentation Index > Fetch the complete documentation index at: https://docs.unifygtm.com/llms.txt > Use this file to discover all available pages before exploring further. # How bidirectional syncs work > An in-depth explanation of how syncs work between HubSpot and Unify. ## Overview Unify can create and update HubSpot records in response to Play runs and Sequence enrollments. This page summarizes the rules used to determine when and how to create or update records in HubSpot. ## Sync timing HubSpot read syncs are scheduled approximately every 15 minutes. HubSpot association syncs run on a separate daily schedule to refresh relationships between records. When a large number of records change quickly in HubSpot, the next scheduled sync may take longer to complete, which can temporarily delay the appearance of new or updated records in Unify. When a HubSpot field mapping changes, Unify starts a background job to reprocess all records for the affected object. This does not interfere with the regular sync schedule, but it does mean that field mapping updates can take some time to be fully reflected in Unify. ## Overwriting data Unify takes a very conservative approach to overwriting existing HubSpot data in order to make data loss impossible. This approach is based around only a few simple rules: * **Creating new records:** When Unify is creating a new HubSpot record, it will fill in all properties that are enabled for writing in the field mapping or default values. * **Updating existing records:** Unify will inspect whether the field is empty or not. If the field is empty, it will be updated with the new value. If the field already has a value, it will only be updated if it is a Unify-specific field. These fields are prefixed with "Unify" and can be found [here](/reference/integrations/hubspot/field-mappings#available-fields). These rules apply regardless of how the record was originally created (e.g., by Unify or externally). If you're looking for more fine-grained control, you can also limit the permissions granted to the Unify integration user in HubSpot. ## Duplicate prevention Duplicates are strictly prevented within Unify. When creating new HubSpot records, Unify will only ever create one record. If there is already an existing HubSpot record of the same type, Unify will always update it rather than creating a new one. However, duplicates are a common problem in HubSpot and may already exist in your HubSpot instance. In addition, users or other integrations may accidentally create new duplicates over time. In order to accommodate this reality, Unify follows specific rules to ensure predictable behavior when updating duplicated HubSpot records. While Unify cannot de-duplicate your HubSpot, it will *never* make the problem worse, and in some situations may be able to help clarify the “source of truth” record. ## Supported objects If there is an existing contact that matches the Unify person being written, it will be updated. Matches are determined based on email address. If there are no matches, a new contact will be created. If there are multiple contacts that match the Unify person, only one will be updated. Specifically, the contact that was most recently modified within HubSpot is the one that will be updated. HubSpot companies are written by Unify whenever a contact is created or updated. They are also written if a HubSpot sync action runs within a Play that is running on companies. If there is an existing company that matches the Unify company being written, it will be updated. Matches are determined based on the domain of the company website. Domains are normalized, so URLs that redirect to different domains will not result in duplicates. If there are no matches, a new company will be created. If there are multiple companies that match the Unify company, only one will be updated. Specifically, the company that was most recently modified within HubSpot will be updated. Unify does not update existing email messages in HubSpot; only emails sent through Unify or received in response to those emails will be written to HubSpot. Email messages are written to HubSpot for emails sent as a part of Unify Sequences and for replies to those emails. The corresponding person must already exist in HubSpot as a contact, or a Play action must sync the person to HubSpot before the email activity is written. Unify writes eligible Unify tasks to HubSpot as HubSpot tasks. This includes ready or completed non-email tasks, such as phone-call and action-item tasks, and completed email or reply tasks. Unify does not write email tasks to HubSpot before they are complete because completing an email task in HubSpot can imply that the email has been sent, while email sending is controlled in Unify. HubSpot tasks are associated with the corresponding HubSpot contact. If the corresponding person does not already exist as a HubSpot contact, Unify must sync the person to HubSpot before the task can be written. Task sync uses HubSpot's standard task properties instead of custom field mappings. Unify writes values such as the task subject, body, task type, status, priority, timestamp, and owner. When a task is deleted in Unify, Unify archives the matching task in HubSpot. HubSpot-to-Unify task sync is limited to completion updates. When a non-email HubSpot task that was written by Unify is marked complete in HubSpot, Unify marks the corresponding task complete in Unify. Email tasks are excluded from this completion sync. HubSpot deals are read into Unify as opportunity records. They can be associated with companies and people when the HubSpot deal has the corresponding company or contact association. Deal data is primarily CRM-owned. Unify uses it for record context, filters, and exclusions, such as excluding companies with open opportunities from outbound Plays. Unify does not create or update HubSpot deals itself.