Analytics news and expert advice every Thursday.
5 Reasons To Migrate To Universal Analytics Sooner Rather Than Later
“Universal Analytics is the new operating standard for Google Analytics. All accounts will soon be required to use Universal Analytics.” — Google
Is the fear, uncertainty and doubt setting in yet?
Well, no reason to panic quite yet, but before you have to switch, let’s talk about why you want to switch sooner rather than later.
Universal Analytics (UA) is still in beta for now, but at Cardinal Path we’re lucky to have been working with UA as a Google Analytics Certified Partner since its very early stages. And we can tell you it’s come a long way; in all but a handful of cases it’s ready to go.
Before we get into why you should probably hurry up and get on the UA bandwagon, let’s take a quick look at the two step upgrade process.
You’re going to log into your GA account and, in the admin section, you’ll have the choice to transfer your existing classic GA web property to UA.
You can do this at any time, and GA will still keep on processing your old code; but, please note that if you’re currently using dc.js (or have in the past) to enable GA features like Remarketing, Google Display Network Impression Reporting, DoubleClick Campaign Manager Integration, and the Google Analytics Demographics and Interest Reports, you won’t see this option.
That’s a good thing. These aren’t quite ready yet in UA, and if you fall into this category, you’ll want to hold off just a bit longer.
Eventually, you’re going to have to change your tracking code from the old ga.js to the new analytics.js. I know, it’s not good news. But it’s not that bad. And, this is going to give you a great excuse to rip out all that old, disparate code from wherever it’s been collecting over the years and upgrade to a tag management solution.
There are many out there, but one option is to use Google Tag Manager (GTM) to upgrade to UA. It not only supports Universal Analytics natively, it also helps you manage all your other tags (and bother IT less).
By the way, make sure not to do Step 2 until the GA Interface tells you that it has finished with Step 1, otherwise you risk losing data; also note this transfer can’t be undone. But, upgrading a property will retain all your historical data, so don’t worry about losing anything. (Editor’s note: here’s a more in-depth guide to upgrading.)
OK, so why go through all of this now?
1. Because Eventually You Have To
Remember, the upgrade is a two step process, and this is the important part: if your account isn’t using anything that’s not already supported in UA, Google is going to do step 1 for you, and they won’t necessarily tell you.
If this happens to you, you’ll see a notice in the tracking code page for the web property. Now, don’t freak out – remember, this doesn’t break anything or change anything – your existing code will still work and do all the things it used to do.
Eventually, though, you’ll need to upgrade to the new codebase. Most of the new features that are rolling out at breakneck pace are ONLY coming out for UA – you won’t get them if you’re still on classic GA. And, if you’re using things like custom variables, you’ll have to switch them to the UA equivalent (see my next point).
2. Custom Dimensions (And Metrics)
Many of you remember the olden days of Urchin and early GA when we had a User Defined variable to work with.
It was great – in case GA didn’t capture everything we wanted it to by default, we had an extra variable we could use. Turns out people liked this and there were a lot more things they wanted to track. Enter custom variables, which let GA users track page, session, and visitor level key/value pairs, providing us ways to track almost anything (think “gender:male,” or “loggedin:true,” or “memberLevel:platinum” – sky’s the limit).
The standard free tool gives you 5 of these custom variables, GA Premium users can use up to 50, and life is good. Segmentation and wonderful analysis? Check.
Now, UA takes this a step further with custom dimensions (and metrics), and you will have to change your custom variables to these. The bonus: you get 20 of them in the free version, and 200 in GA Premium.
This is also a two-step process: first, you tell GA what your custom data looks like in the admin section of the interface, and then you fire the code, much like you did with GA classic’s custom variables.
Now here’s the coolest part: these new dimensions are now available in all your reports – not just in the custom variable reports of classic GA.
That’s right, if you want to define a visitor-level custom dimension of age or income or member or newsletter subscriber or anything else you can dream up, once configured, it’s available just like any standard dimensions that GA is tracking by default.
3. Dimension Widening
Oooh, this one is cool. Have some more information you want to be able to look at in relation to an existing dimension? Now you can.
Let’s make this real with an example: Page Title or Request URI are both good dimensions that can point us to a piece of content; and, let’s say we have a blog with lots of different authors and content categories.
When we’re analyzing our posts to see what we should be writing more or less of, we might want to take a look at more than just each individual page.
What if we “widened” the page dimension with additional data like author, category, format, publish date, and more? Now, we can quickly see this additional data and derive insights around the best combinations of all those extra things to drive our content strategy. Nice, huh?
4. User ID
The User ID is in closed beta for the moment, but it’s worth talking about.
Traditionally, tracking users has been sticky with GA, as Personally Identifiable Information (PII) was always a no-no. However, many were able to work around this by assigning anonymous IDs that came from a CRM or backend database as a custom variable that could later be extracted through the GA API and matched up with the PII on back end systems (and according to privacy policies).
UA still doesn’t permit PII tracking in its systems, but it does now let you assign your anonymous User ID if and when you know it. So when someone logs in, clicks on an email you sent them, or does anything else that allows you to programmatically identify that user, you can send that anonymous User ID to UA.
So, why is this exciting? Because if Google Analytics knows a User ID, it can also report on user behavior across sessions and across devices. And, UA has a host of cross-device reports that can show you how your visitors’ path to purchase moves from tablet to smartphone to laptop to phablet to smartwatch to Google Glass and beyond.
5. Measurement Protocol
OK, here’s where it gets fun. The Measurement Protocol allows you to send data to GA from anything that has access to the internet – think point-of-sale systems, kiosks, call centers, handheld barcode scanners, thermostats, coffee pots — the sky’s the limit.
Now, let’s take a simple example: say you want to use all those pretty graphs and features of GA to track how many people walk into your physical store, and the only resource you’ve got is a bored intern that can code a little bit. With a quick script or mobile app, you’ll be firing a hit back to GA every time that bored intern taps a button. Silly and simple, I know. But you get it: you can send these hits from anywhere.
Let’s take a more interesting example. The Measurement Protocol also lets you send additional data that can be tied back to a web session. Why would that be interesting? How about tracking phone calls?
Traditionally, call tracking solutions could fire virtual pageviews back to your GA account, but, because there was no way to tie them to the visitor’s session that led to the call, they just looked like single pageview visits all alone and unattached to any other piece of data in your account.
But now, that phone call can be sent from call tracking providers’ servers and matched to the session that triggered the call, so that a phone call can be analyzed just like any other online conversion action. You can then see things like which campaigns drove the call, what pages in the session led up to the call, and what events (think video views, PDF downloads, etc.) were instrumental in getting that phone to ring. Cool.
There are plenty more reasons to switch, but these are some enticing ones. If you’re still using dc.js to take advantage of GA remarketing and audience data features, then I’ll forgive you if you’re still double tagging or holding out for feature parity; but, for tips on how Universal Analytics will impact your specific Google Analytics implementation, you may want to spend a few minutes with Cardinal Path’s UA migration assessment tool.
One thing for certain is that Universal Analytics is here to stay, and you can start taking advantage of all of this and more by upgrading sooner rather than later.
Some opinions expressed in this article may be those of a guest author and not necessarily Marketing Land. Staff authors are listed here.