Raw-data reports for PBA

Sample reports

• Website events: CSV, XLS
• Website visits: CSV, XLS

Until November 18, 2020, the fields in the reports were selected in Data Locker. Starting November 18, 2020, the field list is set by PBA and can be changed without notice. Consider this when setting up your parsing processes. 

Raw data fields are populated in part from data originating in the mobile app user attribution link and app install event. 

Download field list as CSV

There is no need to select fields in Data Locker. 

The record structure of the report is as follows:

• The record fields are grouped as follows:
  • Event: The install conversion event.
  • Touchpoint: Touchpoints encountered prior to the install conversion.
  • An install conversion consists of one or more touchpoints. Each touchpoint has a record in raw data. The install_id groups the touchpoints of an install together.

Example web to app report:

• Website assisted installs example 

• There is no need to select fields in Data Locker. The report always includes the fields listed. 

Report record structure

• Fields grouped as follows:
  • Event: The conversion event and associated information.
  • Touchpoint: Details of a touchpoint in the path to conversion.
• A conversion consists of one or more touchpoints. In the file, each touchpoint is represented by a record. 

Example Conversion Paths report
Conversion_paths_example_oct20.csv

• PBA reports are written Daily to the Data Locker in h=23 folder for events occurring on a given day.
• Data can't be written retroactively.
• The Data Locker data retention period is 30 days. 

 To enable PBA reports the account admin must set up Data Locker:

1. Go to Integration > Data locker.
2. Select all the reports in the People-Based Attribution section. 
3. Click Save configuration.

To download reports, see accessing data locker reports.

Data locker PBA raw data reports are made available in the h=23 folder. 

Example: t=website_events/dt=2020-11-26/h=23

The data reflects the users most likely to engage with both your mobile app and website.

Combine mobile app and website data.

Solution: 

• Combine reports using customer user ID (CUID)
  
  • Use the same CUID for both mobile and web and combine reports using CUID as the key. 
  • Then, combine reports using the CUID field
• PBA cross-platform user identification:
  • PBA can determine that a web and a mobile user are the same person.
  • If so, then web reports with PBA include a mobile advertising ID: Android advertising ID, iOS IDFA, OAID, or Amazon advertising ID.
  • Then, combine mobile and web reports based on mobile advertising ID.

Use cases for PBA and mobile

Combine mobile and web data to see user activity across platforms and devices. The combined data lets you: 

• Follow the user journey across your brand's digital properties
• Analyze user behavior in each platform
• Consider which platform shows more user engagement
• See which platform produces more conversions
• Measure your efforts at switching users from web to mobile

PBA raw data is available in Data Locker. For a given report, the data is contained in the h=23 folder of report type and consists of multiple compressed files. 

Use this procedure to:

• Copy the files from Data Locker using Cyber-Duck or any other tool for copying S3 files. 
• Combine the files into a single CSV file using Q. 

Copying PBA data from Data Locker

 Before you begin:

• Install and setup Cyber-Duck. 

To copy data using Cyber Duck:

1. Open Cyberduck
   The list of data types available displays.
   

   

2. Double-click on a PBA report type and choose from web_events, web_to_app, or web_touch_points.
   The list of report dates available displays.
3. Right-click on a given date.
4. Select Download to a specific location and complete the location. 
   The files download. 
5. Repeat the above steps to copy data from other report types. Each report type and date must be stored in a different directory on your computer. 

Creating a combined CSV file: 

Before you begin:

Install Q on your computer.

To combine the data in a CSV file using Q:

1. Using the command line/maс terminal, go to the directory containing the data.
2. Use the command that follows to combine the compressed files into a single file having the name pbadata.csv.
   Note: The symbol ~ represents the path/home folder where the Q command is located on your computer. You may need to include the full path to invoke the command. 
   

   ~q "SELECT * from part*.gz" -d , -H -z -O >pbadata.csv

3. Open the file using Excel or any other tool