Upgrading to Universal Analytics Tracking Code – Preparing for Migration
In April 2014 Google Analytics (GA) released its latest tracking code, Universal Analytics (UA), which has brought a host of new features and functionality to its web analytics platform. While many websites were quick to migrate their tracking code to the latest version, many websites still run on the old asynchronous code which will be deprecated at some point in the not too distant future.
Migrating or upgrading your tracking to Universal Analytics is not a simple switch. Depending on how advanced and detailed your current implementation is, it will require some planning to ensure the data remains accurate throughout. After all, you won’t want to explain to the Board why your numbers are suddenly very different.
Universal Analytics is not only more accurate than its predecessor but also offers a number of benefits and additional features such as UserID (tracking users across different devices), the measurement protocol (sending data to GA from virtually anywhere) and enhanced ecommerce, to name a few. So, there is a lot of additional value to be gained from GA once the migration is completed.
You are probably reading this because you have decided that it is time to upgrade your tracking code to UA. Before you go ahead and just replace the tracking code, we have a few tips for you to ensure the transition goes as smoothly as possible.
Get a measurement plan in place. A tracking code migration is an opportunity to improve your tracking and to add anything you’re currently not tracking. It’s also an opportunity to align your analytics with your business goals to ensure everyone is working towards the same result. You can then use the measurement plan to compare data from the old and the new tracking code too, more about that later. You can find a detailed post on measurement plans as well as a template here.
Check the accuracy of your current implementation. Are you confident that your data is sound? Take some time to compare conversion data with conversions captured in your CRM, for example, and review your reports for signs of self-referrals, spam and bot traffic.
Obvious signs that something isn’t quite right include a bounce rate that is very high or very low, obscure or deep landing pages being recorded for direct traffic, or referrals from your own website. Also check your hostname report domains - they should not be sending traffic to your GA account.
Failing to verify the accuracy of your existing implementation may lead you down the wrong path if you find discrepancies with the new tracking code, so verifying your data before migration is really important
Consider using Google Tag Manager (GTM) for your new implementation. A tag management solution is the most effective way to manage your tracking code and various other marketing and tracking tags on your website. Implementing UA via a tag manager will allow you to make changes and improvements in the future without the need to change code on the site – most changes can simply be made via the tag manager’s interface. For a smooth transition we would always recommend running the old and new code in parallel for a while before ultimately switching the old code off – this process is really straight forward if entirely managed via GTM. If you decide to go down this route, the first step would be to move your existing tracking code (yes, the old one) into GTM. If you need help in using Google tag Manager please get in touch with us, or check out our Google Tag Manager training course
Create a Universal Analytics test property in your GA account to send the new tracking data to. This property should have a single view with no filters set up but it needs to contain all the same goals as your current reporting view
Choose your moment – a tracking code migration requires a few people to pull together so it might be wise not to attempt this just before your business goes into its peak season. Also, try not to migrate the tracking code while major changes are made to your website – such as a platform migration. If they do coincide , the best thing to do would be to migrate the tracking code first, that way you can accurately measure the impact the site migration may have on traffic and performance
Ultimately, you want to make sure that the accuracy of the data is not impacted by anything other than the tracking code implementation itself; this will save you a lot of time when hunting for bugs.
We wrote a very detailed blog post earlier this year on the exact process (including an image of the complete process follow for migration) we follow when migrating tracking code. If you need help with implementing Universal Analytics, please read our blog post.
We’d love to hear any challenges you have encountered when migrating your code or concerns you may have with your upcoming migration. Let us know in the comments or get in touch if you would like Fresh Egg to help you out with gaining insight into your business, or check out our range of analytics training courses.