When creating a data stream in Salesforce Data Cloud, the Event Time Field and Record Modified Field are important settings that affect the accuracy and consistency of data updates.
Setting up these items without a proper understanding of them can lead to problems such as duplicate or missing data.
| Field Name | Categories | Configuration | Example of configuration target |
|
Event Time Field | Engagement | Required | Created date, event date, etc. (date not to be changed in the future) |
|
Record Modified Field | Profile/Engagement/Other | Optional | Last Modified Date (to manage file processing order) |
Point:
The "Event Time Field" plays a role similar to that of the primary key of data, and the "Record Modified Field" determines whether overwriting is allowed or not.
| Update mode | Event Time Field | Record Modified Field |
| Full update | Not applicable | Not applicable |
| Upsert | Applicable | Applicable |
The selected Event Time Field is internally recorded to cdp_sys_PartitionDate. This field is a formula field, and the time is converted to 00:00:00 as follows for DateTime. This process is performed before the time zone conversion, so 9 hours are added later in the case of JST.
Conversion flow (example)
| Processing stage | Value | Remarks |
| On import (UTC) |
2025/1/20 19:00:00 | Event Time Field |
| After formula is applied (UTC) |
2025/1/20 00:00:00 | Reset hours, minutes and seconds |
| After time zone conversion (JST) |
2025/1/20 09:00:00 |
cdp_sys_PartitionDate |
Note:
If the date in cdp_sys_PartitionDate differs even slightly, it is registered as a new record. (This is the reason why it is like a compound primary key.)
Specify date fields that will not change in the future (e.g., creation date, event date, open date, click date).
Once set, they cannot be changed, so thorough consideration must be given before they are set.
The "Record Modified Field" is a field that is used to determine if an overwrite is possible.
The value set on import is managed as a high watermark (maximum value).
| Field | Content |
| Target field type | DateTime only (Date is not allowed) |
| Categories | Profile/Engagement/Other |
| Scope of application | Active only for upsert |
| Condition | Process |
| Larger than the date and time of existing data | Perform overwrite |
| Identical or smaller | Updates are ignored |
Example: If the existing "Record Modified Field" is 2025/11/19 10:25:45
| Import value | Processing result |
|
2025/11/19 10:25:46 or over | Overwritten |
|
2025/11/19 10:25:45 or under | Ignored |
In environments where the order of data processing may be disrupted, it is recommended that "Record Last Modified Date" be set as a record modified item.
If the Event Time Field is different and created as a new record, the "Record Modified Field" is ignored because there is no comparison target.
| Field | Main Role | Operating conditions | Example of recommended settings |
| Event Time Field | Control overwrite/insert as compound primary key | Overwrite if date matches, create new if date mismatches | Created date, event date, etc. |
| Record Modified Field | Controls whether overwriting is allowed or not | Update only if greater than existing date and time | Last Modified Date, etc. |
Correctly configuring these items enables maintaining data integrity and optimizing Data Cloud processing accuracy and performance.
Contributor: Nobuyuki Watanabe | Trailblazer Community Forum Ambassador
Serving as a Marketing Cloud Engagement consultant at FPT Japan Holdings Group. He is working to optimize customer communications using Marketing Cloud Engagement by supporting scenario design, data design, campaign operations, and personalization measures, with a focus on email, LINE, and PUSH.
The posts represent the views of the author who submitted them alone and do not necessarily reflect the position of Salesforce. Regardless of whether or not associated with Salesforce, posts may contain or constitute products, services, information, data, content or other materials provided by third parties ("Third Party Materials"). Salesforce does not control, endorse, or guarantee any Third Party Materials (including their accuracy, validity, timeliness, completeness, reliability, integrity, quality, legality, usefulness, security, or applicable intellectual property rights) and assumes no responsibility therefor. Posts provided in response to questions posted via message boards or forums, or that claim to answer questions, including questions about Salesforce or programs, are provided for general informational purposes and should not be relied upon as answers to specific questions not (even if the answer is marked as the "best" answer or similarly qualified). You should always contact Salesforce support for answers to your specific questions. Salesforce has no control over the content of posts and assumes no responsibility for the use or misuse (including distribution) of the content of posts by third parties.
If you have any questions, please take advantage of the knowledge of our Trailblazer community.
https://trailhead.salesforce.com/trailblazer-community/feed
005227575

We use three kinds of cookies on our websites: required, functional, and advertising. You can choose whether functional and advertising cookies apply. Click on the different cookie categories to find out more about each category and to change the default settings.
Privacy Statement
Required cookies are necessary for basic website functionality. Some examples include: session cookies needed to transmit the website, authentication cookies, and security cookies.
Functional cookies enhance functions, performance, and services on the website. Some examples include: cookies used to analyze site traffic, cookies used for market research, and cookies used to display advertising that is not directed to a particular individual.
Advertising cookies track activity across websites in order to understand a viewer’s interests, and direct them specific marketing. Some examples include: cookies used for remarketing, or interest-based advertising.