AppsFlyer Data Locker

Introduction

AppsFlyer offers a new solution for exporting entire raw data on a daily basis. This data can be processed by customer scripts and imported into their BI systems or be available for pulling when required.  

 Note

This is a premium feature which is an add-on on top of packages. For specific pricing please contact your Customer Success Manager.

For customers with large amounts of data, Data Locker serves as an alternative to exporting data via Push API. Once the Data Locker is configured, AppsFlyer automatically creates a repository for the data and makes it available for the customer’s convenience.

Benefits

  • There is no need to set up any servers on your end
  • Reliable: no data loss since the data is stored in AWS
  • More data: get much more data such as Organic Installs and In-App Events, Sessions, Clicks and Impressions
  • Pull data when required

Required Steps

1. Select the Data Locker option under Integrations (this is only available to account owners).

2. Select the relevant applications for which data is to be stored.

3. Configure the required data: you can choose to receive Non-Organic, Organic and Retargeting data.

The data can include Installs, In-App Events and Uninstalls. In addition, for select customers, we also have the ability to send Sessions, Clicks and Impressions.

4. Select fields (optional)
Click on the the box titled Select Fields to pick the fields required on your data locker reports. Check the box next to any required field to include it in your reports.
If you skip this step or manually select All Fields your reports shall include all possible fields.

 Note

AppsFlyer reserves the right to add new fields without prior notice. If your parsing process is not built for field additions AppsFlyer recommends to manually select the required fields.  

5. Add recipients (optional)


Add the email or emails (separated by ';') of any team members in your account, that would get daily notice emails when data locker files are generated.

6. Click Save.

Once saved, a dedicated AWS bucket is created. The bucket is only accessible using customer credentials (for security reasons). In addition, all access to the bucket is audited.

 Note

Data Locker impressions support is in beta.

Technical Details

This feature is only visible to account owners.

Data Locker files are based on Raw Data Reports V5 (see: Raw Data Reports V5).

See attached sample file of a CSV file below.

 Technical Notes

  • We run a daily batch job that completes around 9:00 am GMT that populates the bucket with the relevant report.
  • Folder structure is: af-ext-raw-data/<Home Folder>/account/<event-type>-<date YYYY-MM-dd>
  • The Home Folder is the Home Folder found in the Credential Details window (see screenshot above).

For example for the date 2016-08-12 the relevant report appears under:

s3://af-ext-raw-data/support-appsflyer-com/account/installs-2016-08-11/

** assuming the home folder is support-appsflyer-com

  • Each file includes all of the selected apps in the Data Locker configuration. apps registered to the account at the time of the report job initialization (relating to both Organic and Non-Organic installs).
  • The report files under those folders are zipped with gz. (to make the download process efficient at the client end).
  • The actual data file is CSV but its name is "part-00000" (with no file ext)
  • Any file generated on S3 is available for 30 days.
  • In each folder, the data is split between many files to avoid files that are too heavy. Depending on the type of data exported, each folder contains between 50 to 1000 files (this number can change without notice). Files are named `part-00000`, `part-00001`, etc
     
    Each file has a header row.
     
    In each folder, after the files are successfully generated, an empty file called `_SUCCESS` is generated. We recommend checking if this file exists before processing the data.
     
    The separator is the coma `,` ; values containing a coma are contained between double quotes `"`, example `"iPhone6,1"`
     
    Each folder is available for 30 days, after which it is deleted.

Accessing the Data

The data can be accessed in two different ways:

AppsFlyer generates an AWS principle (ARN in Amazon terms) under our account and generate AWSKey and AWSSecret credentials for that principle. We then set a policy to allow that principle to both browse and retrieve files from the bucket.  Once you retrieve the credentials from the Credential Details section you can access the bucket using any AWS command line tool as well as most FTP clients.

 Note

We recommend using CLI (Command Line Interface) on Mac and Linux systems for accessing the data. On windows we recommend using Amazon's S3 Browser, but not DragonDisk, which is not fully supported.

 

Example

An AWS GUI Client example:

For AWS Cli Tools, perform the following steps:

Open Terminal at your HOME folder (use the Bucket Name as the Home Folder):

  1. Enter aws configure
  2. Enter your AWS Access Key
  3. Enter your AWS Secret Key
  4. Enter eu-west-1
  5. Click Enter (None)

Now to list your files under your bucket:

'aws s3 ls s3://af-ext-raw-data/<<home-folder>>/'

** valid for LINUX base OS

 Note

The data in the this feature is GMT based and is activity-based for a GMT day (24 hours).

Available Data

For each file, the following information is available:

Folder Description
clicks Clicks
impressions Impressions
inapp In-App Events
inappreattrReport In-App Events resulting from re-attributions
inappretargets In-App Events resulting from re-engagement
installs Installs
reattrReport Re-Attributions Report
retargets Re-Engagement
sessions Sessions
launches App sessions
uninstalls Uninstalls
Was this article helpful?
0 out of 0 found this helpful