Upgrad­ing Your Website’s Track­ing to UA:

The follow­ing content is for exist­ing users of Google Analyt­ics (GA) using the ga.js library that would like to trans­fer your current Google Analyt­ics prop­erty to Univer­sal Analyt­ics. If you do not already have a Google Analyt­ics account, click here to read how to create a Univer­sal Analyt­ics account from scratch.

Switch­ing or upgrad­ing prop­er­ties in your current GA account is a multi-step process. The time and resources required to upgrade your account and fully utilize Univer­sal Analyt­ics will vary depend­ing on a number of factors, includ­ing the size and scope of your current imple­men­ta­tion and time and resource constraints you may have, among others.

When should I upgrade to Univer­sal Analyt­ics?

  • All GA prop­er­ties will ulti­mately be migrated to Univer­sal Analyt­ics. The exact time­frame of the manda­tory migra­tion has not yet been announced; however, the deci­sion to switch now or later will depend on a couple differ­ent factors: — Do you use the dc.js JavaScript, instead of the default ga.js JavaScript? — If so, you will need to wait to trans­fer your prop­er­ties until the features of dc.js are supported within analytics.js (the JavaScript library used for Univer­sal Analyt­ics)
  • Features of dc.js that are not yet supported are: — GA Remar­ket­ing
  • GA Demo­graph­ics reports (age, gender, inter­ests, etc.)
  • Google Display network (GDN) Impres­sion Report­ing
  • DoubleClick Campaign Manager Inte­gra­tion
  • Would you like to view your historic and future data in one prop­erty? — This ques­tion may seem like a trivial one at first, but Univer­sal Analyt­ics gives you the ability to change how some aspects of your data is captured and processed. These changes can make future data analy­sis less accu­rate depend­ing on updates you may make to recog­nized organic search engines, session or campaign timeout para­me­ters, etc.
  • If you plan on updat­ing the processing/classification rules for Univer­sal Analyt­ics, it may be helpful to either create an all new Univer­sal Analyt­ics prop­erty within your GA account, or to create a well detailed docu­men­ta­tion guide on the changes that have been made and what metrics can be accu­rately compared to one another over time.
  • Are you a GA Premium customer? — Currently, the SLAs that apply to GA Premium customers are not supported with Univer­sal Analyt­ics. If you are a GA Premium customer and would like to retain your SLAs, you will need to wait until a later phase in the migra­tion process when the SLAs are supported.

Step 1: Trans­fer your prop­erty to Univer­sal Analyt­ics in the GA inter­face

For this step you will need to be an admin for the prop­erty that you intend to upgrade. If you are not an admin, please forward this link to an admin for your GA account/property and let them know you would like to begin the process.

  • Sign in to your GA account and navi­gate to the admin screen of the account and prop­erty you would like to upgrade. Here is a screen­shot of what you should see for a prop­erty that has not been upgraded:
  • Under the Prop­erty drop­down you will see a link titled ‘Univer­sal Analyt­ics Upgrade’.  Click this link and your screen should display some infor­ma­tion on the bene­fits of upgrad­ing to Univer­sal Analyt­ics, as well as a button near the bottom that says ‘Trans­fer’.
  • Click this button and you should see a modal window speak­ing about the poten­tial limi­ta­tions of upgrad­ing to Univer­sal Analyt­ics (see the ‘When should I upgrade to Univer­sal Analyt­ics?’ section below for more infor­ma­tion on these items). Once you have read this infor­ma­tion click the Trans­fer button again. This will start your trans­fer.
  • Trans­fer­ring your prop­erty to Univer­sal Analyt­ics will take between 24 – 48 hours. The next step is to update the track­ing code for your prop­erty.

Step 2: Updat­ing Your Property’s Track­ing Code

Upgrad­ing your prop­erty to Univer­sal Analyt­ics in GA was a pretty easy step that should only take a few minutes. Once your prop­erty has been migrated within the GA inter­face, now comes the more resource inten­sive part of the upgrade process.

In order to utilize the full bene­fits of Univer­sal Analyt­ics, you will need to trans­fer all of your track­ing codes from using the legacy methods to the methods used by the analytics.js JavaScript library. Depend­ing on the size of your site, the resources you have avail­able, and the timing require­ments you have for upgrad­ing to Univer­sal Analyt­ics, there are two options for how you can proceed:

  • Option 1: Use Airlock to help with the migra­tion — Airlock is a free, open-source product built by Search Discovery for a variety of differ­ent users. If you don’t have the resources to retag your entire site, need to use certain aspects of Univer­sal Analyt­ics but don’t have the time to retag your entire site, or want to upgrade but don’t neces­sar­ily have the tech­ni­cal knowl­edge for your track­ing codes to be updated, Airlock is a great near term solu­tion that will get you using the full bene­fits of Univer­sal Analyt­ics in a very short time frame. Click the follow­ing link to learn more infor­ma­tion on how to install and use Airlock.
  • Long term, we do recom­mend that you either retag your website using inter­nal resources, deploy a tag manage­ment system that works with Univer­sal Analyt­ics track­ing, or by contact­ing a digital analyt­ics consul­tant to help guide you through the process.
  • Option 2: Retag your entire website with updated Univer­sal Analyt­ics track­ing codes. — This option is the preferred method when migrat­ing to Univer­sal Analyt­ics. In order to get the most benefit from this step, we recom­mend devel­op­ing a digital analyt­ics track­ing frame­work and method­ol­ogy for how data is to be captured. It is also bene­fi­cial to create well-detailed docu­men­ta­tion on the differ­ent track­ing require­ments you have and the corre­spond­ing track­ing codes so you will have this infor­ma­tion for future GA updates.
  • To retag your current imple­men­ta­tion there are a number of differ­ent modi­fi­ca­tions that you may need. Below, we will cover the migra­tion of the most common features used. For a more robust list of the differ­ent track­ing code modi­fi­ca­tions you may need, visit this GA Devel­op­ers article for addi­tional infor­ma­tion. — The main track­ing snippet. For every page that you track you should see some­thing similar to the follow­ing code on each page:

When updat­ing to Univer­sal Analyt­ics, all aspects of this snippet will need to be replaced to use the follow­ing code, with excep­tion to the prop­erty ID used in the above _setAccount method:

If your current GA imple­men­ta­tion uses a customized track­ing snippet for things like setDo­main­Name, please visit this GA Devel­op­ers article for more infor­ma­tion.
Event Track­ing. For any events recorded in GA, such as PDF down­loads, outbound links, etc., the syntax of the event track­ing code will need to be updated. Your current event imple­men­ta­tions will look some­thing like the follow­ing: _gaq.push([‘
track­Event’, ‘cate­gory’, ‘action’, ‘labeloptional’, valueoption­al­Num­ber, noninteraction_optionalBoolean’]);

When updat­ing the track­ing codes, you will want to use the same values for cate­gory, action, label, etc. as you already have imple­mented, but you will want to update to the follow­ing syntax:

ga(‘send’, ‘event’, ‘cate­gory’, ‘action’, ‘labeloptional’, valueoption­al­Num­ber, {‘nonInteraction_optional’: 1});

  • In the UA event code snippet above, if the nonIt­er­ac­tion para­me­ter is set to 1, this event will not affect the bounce rate of your website when it fires. If you would like this event to affect the bounce rate, you can leave this para­me­ter off and it will default to a value of 0.
  • Custom Vari­ables. Custom vari­ables no longer exist in the analytics.js library and have been replaced by custom dimen­sions. Custom dimen­sions and custom vari­ables are very similar to one another, except custom dimen­sions must first be created and defined in your GA admin inter­face. You also have more custom dimen­sions avail­able to you (20 vs. 5 for GA Stan­dard and 200 vs. 50 for GA Premium users).To create a custom dimen­sion in your GA admin inter­face, click on the custom dimen­sions tab in the prop­erty section, and then click on the custom dimen­sions link from the expanded section.Click the ‘New Custom Dimen­sion’ button, provide a name, and the scope that you want the custom dimen­sion to have. If you are updat­ing from a legacy imple­men­ta­tion of GA, it is recom­mended that you keep the same naming conven­tions, index, and scope for each custom vari­able that is currently in place.Once your custom dimen­sions are created within the GA admin inter­face, the syntax to use for custom dimen­sions is: ga(‘set’, ‘dimen­sionX’, ‘customdimen­sionvalue’);

  • In the above syntax for Univer­sal Analyt­ics, ‘X’ in ‘dimen­sionX’ is the index number for the custom dimen­sion you are setting. The scope and name of the custom dimen­sion will be matched to the index and value during data process­ing.

  • Virtual Pageviews. Virtual pageviews are prob­a­bly the simplest of the track­ing codes to update syntax for Univer­sal Analyt­ics. Below are ga.js and analytics.js code snip­pets: ga.js virtual pageview code snippet gaq.push([‘track­Pageview’, ‘page path’);

analytics.js virtual pageview code snippet ga(‘send’, ‘pageview’, ‘page path’);

Once your track­ing codes are updated on your website your Univer­sal Analyt­ics trans­fer process will be complete. You are now ready to discover the even more power­ful capa­bil­i­ties of Univer­sal Analyt­ics and start explor­ing its amazing features.

Stuck on a part of the trans­fer process? Need help ensur­ing a smooth tran­si­tion? Contact us to use our analyt­ics consult­ing services for help with a smooth tran­si­tion.