Records Under Management
Records Under Management (RUM) for a particular dataset is the total number of unique records in that dataset. This is calculated as the number of records with unique dataset keys with both identifier data and segment data across your active imports.
Records Under Management (RUM) for a particular dataset is the total number of unique records in that dataset. This is calculated as the number of records with unique dataset keys with both identifier data and segment data across your active imports.
When LiveRamp calculates your monthly RUM for billing, we take the average number of RUM throughout the month. Your total RUM is calculated by adding up the RUM for each of your individual datasets (shown below as "audiences").

Note
For some RUM customers, each different input identifier category incurs a different usage rate, according to contract terms. These input identifier categories differ, depending on whether the dataset is an Activation workflow dataset or a Measurement Enablement workflow dataset. See "RUM Input Identifier Categories" below for more information.
You can choose to have all older files automatically deleted from a dataset once they reach a certain age (this is one way to manage the size of a dataset and to reduce costs if you're using the RUM billing metric). For more information, see "Automatically Delete Older Files".
LiveRamp calculates the number of unique records in a dataset every day. At the end of each month, the daily calculations are averaged to get the RUM for that month.
For example, if you have 0 records for the first 15 days of a 30-day month, and 800,000 unique records for the final 15 days of that month, the average (and thus the RUM) for that month would be 400,000.
If those records were left there (nothing added, nothing removed) for the entirety of the next month, the RUM would be 800,000 for the next month.
See "Viewing Usage" for more information on viewing your monthly usage.
Caution
Are records deduplicated across the entire account? For Records Under Management, records are only deduplicated within each dataset by dataset key. They are not deduplicated across different a datasets.
Note
See the following articles for more information on the various elements that contribute to RUM:
How Records Become Unique Records
Below is an example of how LiveRamp counts unique records in your datasets, which involves both deduplicating records and removing records that don’t have both identifier data and segment data from that count.
For example, take a dataset that contains these five rows of data:

Ann Doe will not be registered as a unique record because she doesn’t have a segment data field associated with her dataset key. So five rows become four records:

Both John Doe and Sue Lee have duplicate records. However, John Doe has the same email as the dataset key, whereas Sue Lee has different emails, so John Doe’s records are deduplicated but Sue Lee’s records are not. Therefore, four records end up as three unique records:

RUM Input Identifier Categories
For some RUM customers, each different input identifier category incurs a different usage rate, according to contract terms. These input identifier categories differ, depending on whether the dataset is an Activation workflow dataset or a Measurement Enablement workflow dataset.
For Activation workflow datasets, there are five categories of input identifiers:
For Measurement Enablement workflow datasets, there are three categories of input identifiers:
Offline identifiers (such as PII)
Online identifiers (such as mobile device IDs, cookies, or CIDs)
RampIDs
RUM FAQs
Are records deduplicated?
Records are deduplicated using a dataset key within a dataset, however, records are not deduplicated across multiple datasets.
Does LiveRamp count unique records that LiveRamp cannot match to a RampID?
Yes, even if we we do not match a record to a RampID at the given time we still count it against your RUM allotment. It’s likely that we will be able to match a RampID to that record at a later date.
Where can I see my monthly usage?
You can view your monthly usage information in Connect by clicking Company Settings and then selecting the Usage Report tab.
What happens to my unused RUM at the end of the month?
Any unused RUM in a given month will expire at the end of the month and will not roll over to the next month.
What happens if I use more RUM than my monthly allotment?
For almost all cases, overages will result in overage charges based on a CPM. Reach out to your LiveRamp rep for details.
What are recommendations to avoid overages?
If you are frequently surpassing your monthly RUM allotment, we recommend that you speak to your LiveRamp rep about increasing RUM minimums.
Otherwise, we advise that you clean out your data often by deleting files or consider full refreshes to update your datasets and remove old data.
RUM is an average, so the earlier the dataset record count is reduced in a given month, the more the current RUM usage will drop as the lower count gains higher weight as the month progresses.
Note
There are implications for full refreshes, so work with your LiveRamp rep to see if full refreshes are right for you.
Deleting fields (aka segments) will not reduce the RUM for that dataset. The records themselves must be deleted by deleting the file or performing a full refresh.
Will deleting built segments reduce RUM counts?
No, deleting built segments will not reduce RUM counts because RUM is based on the unique records within your datasets, not built segments created from those datasets.