Show Menu
TOPICS×

About data file and data sources for customer attributes

Data file requirements and multiple data sources for uploading customer attributes to the Experience Cloud.
You will need access to CRM or similar data from your enterprise. The data you upload to the Experience Cloud must be a .csv file. If you upload via FTP or sFTP, you also upload a .fin file.
Customer Attributes is designed to handle a few files per day. To mitigate the issue of having a large number of small files delaying processing, files sent within 30 minutes of a previous batch from the same organization are routed to a lower-priority queue.

Allowed file types and naming requirements

File Type Description
.csv
A comma-separated values file (such as one created in Excel). This is the file that contains the customer attribute data.
Naming requirements: Ensure that file name extensions do not contain white spaces.
.fin
(Required) The .fin file tells the system that you are finished uploading data. The name of the .fin file must match the name of the .csv file.
Adobe recommends creating an empty text file with a .fin extension. An empty file saves space and upload time.
Note: Renaming a .fin file is not allowed after it is uploaded. The .fin file must be uploaded separately and cannot be a renamed, previously uploaded file.
After you upload the .fin file in the customer attributes FTP, the system retrieves data quickly (within one minute). This differs from other Adobe FTP-based systems, which pick up data less frequently (around once per hour).
The .fin file is not required when using the drag-and-drop upload method.
.gz or .zip
.gz (gzip) or .zip - for compressed files. A .zip file cannot contain more than one file in the archive.
Naming requirements: The name of the .zip or .gz should match the name of the .csv . For example, if your .csv file is crm_small.csv , the .zip file should be crm_small.csv.zip .
The .fin file must match the .csv.

Requirements for the attribute data files

Example CSV
The CSV file must adhere to the following format:
Sample CSV:
The same file viewed in a text editor:
Guidelines
Item Description
Drag-and-drop
The drag-and-drop file should be less than 100 megabytes.
The .fin file is not required when using the drag-and-drop upload method.
Customer ID column
The first column must be a unique customer ID. The ID used should correspond to the ID that is being passed to the Experience Cloud ID Service.
For Analytics, the ID being stored in a prop or eVar.
For Target, the setCustomerID value. (See Analytics & Target - synching the customer ID )
This customer ID is the unique identifier your CRM uses for each person in your database. The remaining columns are attributes that come from your CRM. You will choose how many attributes to upload.
Friendly, readable names are recommended for the column headings, but not required. When you validate the schema after upload, you can map friendly names to the uploaded rows and columns.
About Customer IDs
Typically, an enterprise uses a customer ID from a CRM system. This ID is set using the setCustomerIDs call when a person logs in. This ID is also used as the key in the CRM file that is uploaded to the Experience Cloud. An Alias ID is a friendly name for a data store in Audience Manager, where the alias data is stored. The system sends aliases to this data store (via setCustomerIDs). The CRM file is applied to the data in that data store.
For setCustomerIDs information, see Customer IDs and Authentication States .
Subsequent headers and columns
Subsequent headers should represent the name of each attribute.
These columns should contain customer attributes that come from the CRM.
Attribute limits
You can upload hundreds of .csv columns to the customer attribute service in the Experience Cloud. However, when configuring subscriptions and selecting attributes, the following limits apply depending on the solutions you own:
  • Analytics Standard : 3 total
  • Analytics Premium : 200 per report suite
  • Target Standard: 5
  • Target Premium: 200
Row limits
There is no known limit to the number of rows.
Column limits
For practicality, limit the number of columns to around 200.
Character limits
When creating an Analytics subscription, field lengths for the uploaded files are truncated to 255.
FTP Guidelines and size limitations
  • Maximum file size limit for FTP is 4 GB for each upload.
  • Minimum file size limit for 10 mb for each upload.
  • You can upload one file every half hour.
  • You should drop your .csv (and .fin ) file in the root folder of the FTP site.
Important: The total allowed space for the FTP account is 40 GB. It is your responsibility to delete processed files.
File requirements
Each attribute source should contain the same number of comma-separated fields.
Fields containing a line-break, double-quote, or commas must be quoted.
Double-quote characters in a field must be escaped using a backslash (\).
Blank columns are stored as null .
Multiple files
When uploading customer attribute data, if you have several files you want to upload in rapid succession, and especially if the files are large, make certain that the previous file has been processed before uploading the next file. You can monitor this by checking when the previous file has been moved to the processed or failed folder within your customer attributes FTP account.
Breaking a large file into smaller files and submitting them in rapid succession may actually slow down processing unless you can ensure each file is completely processed before submitting the next.
Character Encoding
For Japan, UTF-8 is mandatory.
Historical data
Customer attributes are tied to the underlying visitor profile in Analytics. As such, customer attributes are associated with the visitor for the entire life of that visitor profile in Analytics. This includes behavior that occurred before the customer logged in for the first time.
If you use the data warehouse backfill method, the data is tied to a post_visid_high/low that is based on the Analytics ID (AID). If you are using the Experience Cloud ID service, the data is tied to a post_visid_high/low that is based on Experience Cloud ID (MID).
Data feeds
Customer attributes are not available in data feeds.

Leveraging multiple data sources

When creating, modifying, or deleting customer attribute sources, there is a delay around one hour before IDs begin synchronizing with the new data source.
The Alias ID for each customer attribute source must be unique. If you have multiple data sources that leverage the same ID, they should be set up as follows:
In VisitorAPI.js or the Experience Cloud ID tool in dynamic tag management:
Set two customer IDs that will correspond to the appropriate data sources:
Visitor.setCustomerIDs({ 
     "ds_id1”:"123456", 
     "ds_id2":"123456" 
});

(See Customer IDs and Authentication States for more information.)
In the Experience Cloud > People > Customer Attributes :
Create two customer attributes sources using unique alias IDs corresponding to the customer IDs above. Using this method allows the same reference ID to be sent to multiple customer attribute sources