When you run a mass update, Rhythm calculates the new field value for each selected record. Then Rhythm asks the application that owns the record to save that value. Use this reference to understand how Rhythm calculates values and what the preview shows.
How to Read This Reference
- Current value means the value on the record before the update. New value means the value the rule produces.
- "Empty" means the record has no value for that field. The values
false,0, and empty text still count as real values. - The application that owns the record checks each new value when Rhythm sends it. If the application rejects a value, such as text in a date field or a choice that does not exist, only that record gets an error. Rhythm does not check ahead of time whether the value matches the field type, so a clean preview is helpful but not a guarantee. Preview also does not always catch a value that is wrong only because of capitalization, so some records can still fail during execution until the case matches exactly.
Ways to Give a New Value
| Method | Where you use it | What each record gets | Notes |
|---|---|---|---|
| Fixed value | Enter the new value (or, use an expression): on the values step | Every selected record gets the same value | If you leave the value empty, Rhythm uses an empty new value without warning, which clears the field on every record that currently has a value |
| Yes/No checkbox | Check to set Yes, leave unchecked for No | Yes when checked, No when unchecked | Rhythm shows this option for true/false fields and Yes/No custom fields. If you do not change it, Rhythm sets No instead of leaving the field unchanged. Some screens show these values as TRUE and FALSE rather than Yes and No; TRUE means checked/Yes, and FALSE means unchecked/No. A Yes/No custom field stores only this value. Checking it does not trigger or control any other mass update behavior, though you can still use the field to filter or query records |
| Expression | Enter the expression (or, specify a value): | A value calculated from that record's own fields, such as first name plus last name | If the expression returns a number, Rhythm turns the result into text. If the expression fails for a record, Rhythm marks that record Error and skips it |
| Expression only | Fields marked on the values step with This is a complex field type and can only be updated via an expression. | The value an expression produces | For fields marked this way, you can set lists and structured fields in the wizard only with an expression |
| File cell | A column mapped to the field in a file-driven update | The value from that record's own row | An empty cell gives Rhythm an empty new value, which clears the field when the record currently has a value. A cell that reads [ignore], with any capitalization and spaces allowed, leaves the field unchanged for that record |
If clearing a value from Bulk Changes › Mass Update on a query or search results list does not work as expected, try the same clear from System & Security › Mass Updates instead. Some customers have found that path works when the query-based path does not.
Changing Multi-Select and Other List Fields
A multi-select picklist or another list field stores more than one value. By default, a new value replaces the whole list. To add or remove specific values instead, use this bracket syntax as a fixed value, an expression result, or a file cell:
| You write | Meaning | Example on a record holding ["X"] | Result |
|---|---|---|---|
[A,B] | Replace the whole list | [A,B] | ["A","B"] |
+[A] | Add to the start | +[A] | ["A","X"] |
~[A] | Add to the end | ~[A] | ["X","A"] |
-[X] | Remove each named value if it is present | -[X] on ["X","Y"] | ["Y"] |
'[A,B] | Treat the brackets as ordinary text instead of list syntax | '[A,B] | The text [A,B] |
(d|)[A,B|C] | Use a different separator, here |, when values contain commas. You can combine it with an operator, such as +(d|)[A|B] | (d|)[A,B|C] | ["A,B","C"] |
A (no brackets) | Replace the list with one value | A | ["A"] |
Rules for list values:
- Rhythm splits values on commas, or on your chosen separator. It then trims each value, drops empty entries, and removes surrounding double quotes.
- When the field is empty,
+and~work like a plain replace, and-does nothing. - Removal matches values exactly, including capitalization. If the value you name is not present, Rhythm ignores it.
- If you use any other character before the bracket, Rhythm stops that field with unsupported array operator.
- Use square brackets only; curly braces are not list syntax.
- Relationship lists, such as contact roles, store IDs. Write role IDs, not role names, because Rhythm does not convert names to IDs. Status Reason works the same way outside list syntax: give the underlying Status Reason ID, not the name shown on screen, or the update does not take effect.
- In a file-driven update, put the bracket-encoded value in one cell for that record's row. A file that lists one display value per row, instead of the encoded value in one cell, is not a supported format and can fail with an unclear error rather than a helpful validation message. This bracket syntax is confirmed for fields such as contact roles. Confirm support for a given field or record type, such as a membership section field, before you rely on it, because coverage can vary.
- A validation error that names a field and says must be an array, such as
category_ids must be an array, means the value was sent in the wrong format for a multi-value field. Use the bracket syntax above; the message does not mean the value itself is invalid.
Example: To remove one retired interest from every record without touching the others, use -[RETIRED-VALUE]. To add a role to everyone, use +[ROLE-ID]. If you write only the new value, Rhythm replaces every record's list with that single value.
Custom Fields
Custom fields appear under Custom Fields on the field step, and you can map them from a file just like built-in fields. Rhythm reads each record's current custom field value directly, so the preview shows the real Old value. If Old is empty for a record you know has a value, you chose a different field from the one that holds that value. This often happens when a similarly named field exists on another record type. If you enter text for a Yes/No custom field that is not a true/false value, the owning application rejects that record with an error. In a file-driven update, use true/false or Yes/No rather than 1/0.
Tip
Rhythm-recognized phone number fields are normalized to a standard format when the new value clearly matches a valid domestic pattern. When a number does not clearly match an expected pattern, Rhythm leaves it as entered rather than guess at a format.
If you use a text field to store a date-like value instead of a Date field, write the full date, such as 08/15/2025, not a partial value such as 08/2025. A partial value can save on the record but fail to appear when you use that field in a query.
Blank and Unchanged Values
| Situation | What happens |
|---|---|
| Current value has no value, and new value has no value | Rhythm skips that field for that record |
| No field on a record has anything to change | Rhythm skips the whole record. The record still shows Ready For Execution in the results and is not counted |
| New value has no value, and current value is present | Rhythm clears the field |
| Empty text replacing empty text | Rhythm sends the value anyway, because empty text is a real value |
| Record changed but ends up identical | The results show No Change for that record |
Numbers That Should Be Text
Spreadsheets often store identifiers and postal codes as numbers. When Rhythm is about to send a new value that is still a number, it turns that value into text if the field's name ends in _id. Rhythm also turns the number into text when the field is a postal code field, the field is the record ID, or the field currently holds text. Other fields keep numbers as numbers.
Rhythm cannot restore leading zeros that the spreadsheet already dropped. A postal code stored as 2134 is sent as the text 2134, not 02134. Format those columns as text in your spreadsheet before you save the file, and check the preview's New column.
This also applies to the identifier column a file-driven mass update uses to match each row to a record. A numeric-only ID, such as an Account ID, can be reformatted by your spreadsheet software the same way a postal code can. If that happens, the affected rows can fail with a not-found or generic matching error instead of updating. Format an identifier column as text before you save the file, especially when the identifier is numeric-only.
Dates from Spreadsheets
Rhythm converts date cells only in .xls and .xlsx files, and only when the spreadsheet stores the cell as a date. Values in .csv files pass through unchanged, and text that only looks like a date also passes through unchanged.
| The cell shows | Rhythm sends | Example when your Rhythm account uses New York time |
|---|---|---|
| A date and a time of day | That date and time in the time zone for your Rhythm account | 2025-08-15 14:30 becomes 2025-08-15T14:30:00-04:00 |
| A date only | Midnight UTC on that date | 2025-08-15 becomes 2025-08-15T00:00:00Z |
Rhythm sends a date-only value as midnight UTC, so some screens in American time zones show it as the evening of the previous day. If dates look one day early, check how the field displays the value before you assume the file was wrong. Your Rhythm account's time zone is set in Account Settings: Address, Time Zone, Email Identity, and Logo.
A date you copy from a query export or another download does not necessarily match what a mass update file expects. Before you use an exported date column as the source for a mass update, check it against the rules above and reformat it if needed.
Where This Applies
These rules apply to mass updates you run from System & Security › Mass Updates and from Bulk Changes › Mass Update on results lists. In a custom data task, each mapped column's value goes into the data the task receives instead of being compared with an existing record's value.