SQL Editor FAQs

A:

The initial ingestion of a new dataset, including configuration and identity resolution, typically takes five business days to become available as a new table in SQL Editor. Subsequent updates to a data feed can occur at a daily cadence.

A:

Second- and third-party data can be ingested using the same process as first-party data or licensed from LiveRamp's Data Marketplace team and delivered to the All Assets section of SQL Editor. Work with your LiveRamp representative to access the needed second- and third-party data. Accessible data assets include first-party, licensed third-party data, and any partner assets to which you have been granted access.

A:

Partner data is never joined into a single table. However, a single view can include "virtually joined" data from multiple partners by applying appropriate data permissions. That way, query results include the data from two or more partners in one view as long each data owner's rules are met. For more information, see "Create a Data Permission".

A:

You can write any valid query you want and SQL Editor will verify access and usage restrictions defined by the owner of the data assets. If a query does not meet data permissions, query thresholds, and other restrictions, an alert will indicate why the query cannot return results. If restrictions are partially met and are defined in a way that permits partial query execution, the permitted results are returned.

Querying against a table that includes a large number of unmatched RampIDs can be problematic if it results in an exponential join. Use the rampid_matched function for these use cases.

All first-party data can be queried at row-level granularity, while partner data can be queried and joined at row-level granularity. Results will be displayed in either row-level granularity or aggregated format, depending on the permissions set by the data owner.

A:

If your query contains a join of 2 or more base tables with rampid columns, make sure to use rampid_matched(rampid) on both sides of the join. Base tables with a rampid column will usually contain a number of unmatched rows. When joining without the extra rampid_matched filter, both sides of the joins with unmatched rows will cause an "exploding join". For example, if each side has 100,000 unmatched rows, the resulting join for just these rows will be an extra 100,000 x 100,000 = 10,000,000,000 rows.

A:

You can sample table rows by specifying a percentage. However, LiveRamp recommends that you instead use the rand SQL function.

A:

Create a "LiveRamp Data Hub" case in the LiveRamp Community portal or contact your LiveRamp representative.