Skip to main content
menu_icon.png

Everything you need to switch from Optimizely Classic to X in one place: See the Optimizely X Web Transition Guide.

x
Optimizely Knowledge Base

Changes to the raw data export process in Optimizely X

 
relevant products:
  • Optimizely X Web Experimentation
  • Optimizely X Web Personalization
  • Optimizely X Web Recommendations

THIS ARTICLE WILL HELP YOU:
  • Migrate your raw data import process

As part of scheduled work to improve the performance and reliability of our Data Export feature for Optimizely X, we are changing the way you access your Optimizely X raw data files.

Action Required

Please modify your raw data import process to use the new AWS bucket name, folder path, and file naming structure according to the migration steps listed below. We will no longer generate raw data in the current AWS bucket after July 14, 2017. 

Migration Steps

  1. Update your raw data import process to use this AWS bucket name: optimizely-export-ng/

  2. Update the read folder to account_id/project_id/2.0/yyyy/mm/dd/experiment_id/
    For example: 123456789/234567890/2.0/2017/04/06/987654321/

  3. Update your process to extract and read the file parts.
    In the new AWS bucket, the raw data is stored in multiple file parts instead of a single file (like in the current bucket). The file parts use the following naming structure:
    experiment_id-filepartnum-yyyy-mm-dd-r-reducernum.gz
    For example: 987654321-0-2017-03-06-r-00062.gz 

Please note the following conventions:

  • The unzipped files do not have any file extension, but use the same tsv format

  • There are up to 30 file parts per experiment per day (filepartnum = 0-29)

Complete instructions for Optimizely’s Data Export are available in our developer docs. If you have questions, please file a support ticket or contact your Customer Success Manager.

FAQ

Why are you making these changes?
The changes are part of a project to enhance the performance and reliability of Data Export for Optimizely X, which we plan to roll out by July 2017.
When do these changes take place?
Raw data for Optimizely X experiments will be available in both current and new AWS buckets till July 14, 2017. After July 14, raw data will only be available in the new optimizely-export-ng bucket.
Should I expect any other changes between now and July 14, 2017?
We do not plan to make any other changes. However, if we do need to modify any of the migration steps, we will inform you by email.
Will I be able to access existing raw data in the current AWS bucket?
Yes. All historical raw data (before July 15, 2017) will continue to be available in the optimizely-export bucket.
Do I have to make any changes to my raw data parsing process?
No. The raw data format remains the same.
Do I have to make any changes to my raw data import process for Optimizely Classic experiments?
No. If you are using data export files for Classic experiments, you can continue using the current AWS bucket (optimizely-export) without any changes.
How can I automatically check when the raw data is available for a given day?
Currently, this is not available. We are working to have this capability available by the July 14, 2017.