Avoiding File Ingestion Failure
There are many possible considerations to keep in mind when creating files for upload, but the cause of file ingestion failure or delay often falls into one of the categories listed below.
To be able to automatically ingest files that you upload for processing (which speeds up and streamlines the ingestion process for your files), we create an automation configuration for that particular audience. Often we will use a file from another of your audiences, or the first file you upload to your new audience, as a "seed" or "template" to create this configuration.
Once this configuration has been enabled, it's very important that subsequent files that are uploaded for that audience stay consistent with the "seed" or "template" configuration. If you upload a file that is not consistent (such as a file with different headers or a different audience key), the file will be paused in the ingestion process until the file is manually mapped to the audience.
For more information on the automation configuration, see "The Ingestion Automation Process for File Uploads".
Note
If you need to upload a file that does not match the initial automation configuration, create a support case so we can make sure the file is ingested correctly.
To avoid ingestion issues, be sure to keep the same file formatting as the original seed file, especially including the following:
The same identifier fields (including having the same audience key field(s) as the original file
The same field names (headers in a column-based file and keys in a key-value file) for identifier fields and for any previously-included segment data fields as the original file
Note
Files uploaded for a particular audience are put into a queue for that audience, so a delay in the processing of a file containing issues will delay all subsequent files that have been uploaded to that audience.
If you upload a file with new segment data fields that weren't in the original file, those fields should upload without an issue. However, you might want to monitor the file upload on the Files page to make sure there are no problems. If the processing pauses or fails, use the appropriate Troubleshooting File Upload Issue quick case to have the Support team investigate the issue.
For examples of properly formatted files (including downloadable files that can be used as templates), as well as files containing common errors, see "File Formatting Examples."
Header Issues
Header issues are the most common cause of file upload failure. The most common header issues are:
A header for an identifier column (including an identifier column that is being used as the audience key) that was in the original file uploaded for that audience is missing, or is named differently (including being capitalized differently).
Caution
When including segment data columns that were in previously-uploaded files, make sure that the headers for those columns also stay consistent. If a header for a segment data column changes, a new field will be created instead of having the data from that column update the existing field.
If you are using the "full refresh" option to update the audience, keep in mind that any previously-uploaded segment data columns need to be included in the file if you want to keep that data in the associated fields. Otherwise, if a previously-uploaded segment data column is not included, the associated field will have all the members removed. See "Ways to Update an Existing Audience" for more information.
The file contains two or more headers that are exactly the same (sometimes extra headers and columns are added when saving from Excel).
Make sure that:
Every column-based file you upload has a header row.
Make sure that the headers are unique and that headers for identifier columns (including identifier columns that are being used as the audience key) and headers for any previously-included segment data columns match the headers in the original file uploaded for that audience.
See "Formatting Column-Based Files" for more information.
Audience Key Issues
Keep in mind that every file uploaded to a particular audience in an Activation workflow must use the same audience key as the original file uploaded for that audience so that we can effectively deduplicate and consolidate all rows within that audience.
The most common audience key issues are:
The audience key is different from the one used for the first file uploaded for that audience.
An identifier that was used originally as the audience key is not included in the file, such as “city” when the audience key is name and postal address.
The header for an audience key column is different from the header used in the original file.
The audience key has a low fill rate. For best results, and to avoid having ingestion processing pause or fail, make sure the audience key has a fill rate of as close to 100% as possible.
Identifier Issues
An identifier is formatted incorrectly, such as including both the first and last name in the same field instead of having separate “first name” and “last name” fields.
See "Formatting Identifiers" for more information.
Delimiter Issues
There is an extra delimiter (such as a comma) in a row, making it appear as though the row has more fields in it than it should. When this happens, our system gets confused while trying to parse your file. Watch out for this, especially when creating CSV files from Excel files.
The data contains the delimiter but is not enclosed in quotation marks. Data containing punctuation characters are at risk of delimiter collision and thus data bleed, where the specified delimiter (such as a comma) also appears as part of the data values. This can cause LiveRamp to interpret data in a particular row as belonging to the wrong field. See "Formatting Guidelines for All Files" for more information.
Raw Field Issues
A "raw field" is a field in a given file that has over 250 distinct values. Most files uploaded to LiveRamp should not contain raw fields unless you have previously coordinated this with your LiveRamp account manager.
The values in raw fields cannot be separated out and managed individually in Connect. Many platforms also cannot accept raw fields.
For more information, see "Raw Field".
New Identifier Field Issues
After the upload of the first file in an audience, subsequent files need to use the same file formatting. All originally-included fields should be included with the exact same headers. If you include an additional identifier field (such as a new email address column in a column-based file) without having LiveRamp first adjust the mapping to include that new field, the file will not process until that mapping has been adjusted.
To avoid delays or processing failures, if you need to add a new identifier field to a subsequent file, first create a support case so that we can adjust the mapping before you upload that file.
Multi-Value Field Issues
The default format for fields in a file for ingestion (columns in a column-based file) is for each field to be a "single-value" field. A single-value field is a field where each consumer record can only be a member of one segment in that field. For example, a file might have a column for the field "Favorite_Pet", where each consumer record can only have one value associated with them (such as "Favorite_Pet=Dog" or "Favorite_Pet=Cat", but not both).
But for some fields you might want to allow each consumer record to have multiple values for that field. For example, you might want to create a field in Connect called "Pets_Owned" where each consumer record might be a member of multiple segments, such as "Pets_Owned=Dog", "Pets_Owned=Cat", and "Pets_Owned=Hamster". This type of field is called a "multi-value field".
For consumers that have multiple values for a particular multi-value field, make sure to create separate rows for each value. For more information on formatting multi-value fields, see "Formatting Guidelines for All Files".
Before you upload the first file to a new audience where the file will contain one or more multi-value fields, use the Set Up Audience for First File Upload (Activation) quick case to create a support case to make sure the audience is configured correctly to process those multi-value fields.
Once the ingestion automation process has been set up for the initial file, make sure to create another support case if you change any of the single-value fields to multi-value fields so that the file ingests properly.
For more information on multi-value fields, see "Multi-Value Fields".