The Raw object data store option provides an alternative to the standard Vault object data store method which can improve performance and consistency if used with an appropriate record set. This option is best used for objects with large numbers of small records, frequent object record creation or editing, and predictable viewing patterns.

Admins should carefully consider whether to use the Raw data store setting for a given object, as raw objects have limitations not encountered with standard Vault objects. Use the below guidelines to help determine whether your object should use the Raw data store option.

The Raw data store option provides improved performance and consistency in cases where:

  • The object has a large number of records, in the millions. Raw objects can have up to one billion records.
  • Records are frequently created or updated, and often through integrations.
  • Users filtering and viewing the object records do so with predictable methods.

The Raw data store option is unlikely to be advantageous in cases where:

  • Object fields and other configuration attributes are expected to change frequently.
  • Admins and users require advanced features of the standard Vault object search, such as faceted field filters or record-level security.
  • Users require a high degree of flexibility in searching records.

Raw Object Limitations

Admins should understand the limitations of raw objects before configuring them.

Users viewing raw object tabs can use the Vault search bar to find records. Raw objects also support case-sensitive record filtering. In order to use filtering on custom fields, an Admin must first define a custom filtering index. Without custom indexes, users may experience slow responses or timeouts when filtering. Additionally, searches on raw objects cannot return more than 1,000 records.


If a raw object has more than 10,000 records, the following modifications require the use of configuration packages or developer tools:

  • Add or remove fields
  • Change the length of any variable-length field, such as Text, Long Text, or Rich Text
  • Create indexes
  • Enable lifecycles
  • Enable or disable object types

Learn more about how to modify raw objects.

Multi-Value Picklists

Raw objects support a maximum of two (2) multi-value picklists.

You cannot use the contains operator in filters on multi-value picklist fields in raw objects.


You can use lookup fields on raw objects within standard and Multi-Pass reports. Reporting functionalities such as conditional fields, grouping results, and sorting are available for these reports. However, filters are not supported for non-searchable lookup fields in standard reports.

Example: Imported Legacy Audit Trail Data

Suppose an organization wants to move a large archive of legacy audit trail data into Vault, and the organization’s database administrator needs to determine whether they should use the High Volume data store option. The database administrator would first evaluate the nature of their data and user needs against the functionality of high volume objects:

  • Data Volume: The volume of the audit trail data is very high. The Raw data store would allow Vault to load data faster than the Standard data store.
  • Update Frequency: Once imported, audit trail data records are unlikely to be updated or created. While this aspect of the data does not take advantage of the frequent update capability of raw objects, there is no negative effect from this attribute.
  • Structured Data: The audit trail data is firmly structured and will not require future changes to fields or other configuration attributes. This data would not conflict with the limitations of the Raw data store.
  • Searching & Filtering: Users searching the imported audit trail data will not require advanced search features. Users typically look for audit data in predictable ways such as filtering by a date range or an Affected ID value. This search behavior fits within the limitations of the Raw data store.

After considering these factors, the database administrator finds that there are several benefits and few or no drawbacks to using the Raw data store option, and decides to move ahead with that option.