Coming Soon: NPR Digital Services and Station Analytics System Moves to Universal Analytics

May 5, 2015

Update: Due to a late breaking bug in final QA, we are moving the transition to December 1 to keep data integrity a priority. 

NPR Digital Services’ Station Analytics System (SAS) relies heavily on Google Analytics, including stations’ Google Analytics data. Google Analytics announced several months ago that they were beginning a transition to the next version of Google Analytics, which they are calling Universal Analytics (UA). As Google describes it, UA is “a set of technological innovations that improve the way data is collected and processed in Google Analytics.” Because of our heavy reliance on and use of Google Analytics, NPR Digital Services has been closely following this rollout process.

As of recently, the project is officially out of Beta, and we at NPR are ready to begin our own transition to the new service. We are pleased to announce that we will officially switch over to UA on November 1, 2015. In order to ensure that we do not have any interruptions in data tracking, this transition will need to be a carefully coordinated and collaborative project between SAS stations and NPR Digital Services. Luckily, there is very minimal work required on the station side.

Why Universal Analytics?

First and foremost, NPR Digital Services is dedicated to preserving station and NPR data integrity. Because Google Analytics will eventually force the UA transition upon all of its clients, and the integrity of station and NPR data during this upgrade is of utmost importance, NPR Digital Services is preparing proactively for the change. With a little preparation and planning, we can be sure that our data is not compromised in any way by the transition.

There are new benefits to using UA as well. The new system gives us access to expanded collection methods and reporting tools, which will allow NPR Digital Services to provide broader insights to the station community. For example, UA allows for the creation of new custom dimensions, which will allow greater tracking flexibility tailored to the public media experience. Dimensions have always been available in Google Analytics. They allow you to see how specific metrics can be tracked and distributed across a particular value (like pageviews (metric) across major cities (dimension). These dimensions will now be customizable so that users can define their own dimensions and metrics, allowing much greater flexibility in tracking the data that is most important to public media.

Timeline

The November 1, 2015 switchover date is not arbitrary.  By choosing the first of a month that also falls on a Sunday, we can be sure that we are cutting over from the old version of Google Analytics to Universal Analytics at the start of a full week and full month. This means that when we are reporting on trended data, we can cleanly view what data came from earlier GA and what data came from UA, and treat it accordingly. This ensures that we are keeping trends intact without having to parse out pieces of weeks or months.

NPR Digital Services will be making changes at exactly 12:01am on November 1, 2015. In order to avoid an interruption in data, SAS stations will need to make the switch to UA on their individual accounts prior to this date. The ideal window in which to make these changes is one month to two days prior to the NPR Digital Services switchover date, meaning October 1, 2015 - October 29, 2015. We will be sending out reminders with detailed instructions as we approach this window of time.

Can’t NPR Digital Services do all of this for me?
YES! We would be happy to make the necessary changes for you in your Google Analytics account.  If you would like NPR Digital Services to make those changes for you, you can grant edit permissions to the NPR Digital Services user on your account. NPR Digital Services has prepared a step-by-step how-to guide to show you how to grant this access in your Google Analytics account. View that guide here.

If you have any questions about this transition or process, please email us at analytics@ds.npr.org.