Is Google’s Universal Analytics A Game Changer?
Google’s Universal Analytics is now available to the public, and some have speculated that it will change the way businesses use data. But, is Universal Analytics (UA) truly a game changer? Many of the benefits promised by UA hinge on two new wrinkles in the platform: 1) the ability to get data into UA from […]
Google’s Universal Analytics is now available to the public, and some have speculated that it will change the way businesses use data. But, is Universal Analytics (UA) truly a game changer? Many of the benefits promised by UA hinge on two new wrinkles in the platform: 1) the ability to get data into UA from any source, and 2) the shift from tracking visits to tracking visitors. Let’s take a closer look at each:
Sending Data To Google
Google’s new “Measurement Protocol” makes it possible to send data into UA from any device that captures and transmits data. (That’s right, any device. The folks at Loves Data even used the RFID in employee badges to track caffeine consumption.) This means that you can add your own data from multiple sources into the UA platform. Here are just a few examples of data that can be sent to UA and incorporated into reports:
- Offline Conversions. Send data to UA about in-store foot traffic and sales from the cash register or repeat sales made over the phone with a sales rep.
- Social Interactions. Send data to UA showing engagement/user interaction in social channels (e.g., Likes on Facebook or +1s on Google Plus). You can also send data about social buttons embedded on your Web properties.
- Games. Create custom metrics such as completions or high scores, and send these directly into the UA platform.
- CRM Data. Send data stored in your CRM system into UA and integrate it with other metrics. For example, send gender data into UA from the CRM system to see pageviews by gender of users who sign into a site or app.
Inherent to the Measurement Protocol is a new parameter called User ID, which leads us to the next big shift in the analytics approach of UA.
From Visits To Visitors
Google Analytics (the traditional version) is based on the Urchin product (acquired by Google in 2005), which was designed to track visits to a website. Each time someone “visits” — an app, a website, etc. — GA initiates a Client ID (CID) that represents an anonymous user. This means that when Jane uses an app on her mobile phone, GA initiates a Client ID and counts Jane’s visit from her mobile phone as one unique visitor. Later, when Jane uses her tablet, yet another Client ID is assigned and GA reports this as two unique visitors. The problem is that GA counts Jane as two unique visitors, not as one visitor who used two devices.
Universal Analytics provides a new parameter for tracking visitors: the User ID (UID). When Jane logs into an app on her mobile phone (using a login to the app), a UID is sent from the auth system to GA. Later, when Jane pulls out her tablet, the same thing happens. With UID, Jane’s multiple visits from different devices are counted accurately as a single unique visitor. (For more on how this works and some great screen shots to illustrate, go to 12:35 in Google’s presentation at I/O.
The User ID feature could be used to align data across multiple platforms such as store, desktop, mobile website, mobile app, call center, catalog, etc., in order to analyze users (as opposed to sessions). Justin Cutroni provides a fascinating use-case scenario of how this might work for a garden supply company.
Is UA Really A Game Changer?
There is no question that the concepts behind UA represent more futuristic thinking about analytics than the traditional version of GA. But, these concepts aren’t completely new.
Other vendors were first out of the box on both data integration and measuring visitors as opposed to visits. (See an interesting comparison of approaches here.) What is new, however, is that UA is free — meaning that sophisticated analytics is now available to everyone.
Barriers To Adoption
Still, a few things could potentially interfere with wide-spread adoption:
- Free Isn’t Really Free. The Google team has gone a long way toward making UA as easy as possible — but, as is often the case with analytics tools, fully utilizing UA requires technical skills and development resources. Smaller players may not have the resources required to fully take advantage of what UA offers.
- The Sign-In Requirement. Full utilization of the really cool features in UA (multi-device measurement, for example) seems to be limited to scenarios in which a visitor signs in and is identified (a sign-in to an app, for example). For organizations that rely heavily on registering users/visitors, UA will be a big step up from traditional GA — but it is hard to say at this point how valuable it will be for those who don’t.
- Presentation. One of our biggest problems in analytics is data silos, so tools that help us to integrate data are sorely needed. And yet, getting data into a system is only the beginning. The big challenge is integrating data in useful ways – and this starts with the sticky problem of how to display and report inconsistent data types and metrics. In the case of UA, Google allows you to create custom metrics for the data sources you pull into UA, but you must record it with an existing data type, like a pageview, event, or e-commerce transaction. (For example, if you are reporting on phone orders, the number of orders could be shown as “Visits.”) I don’t see this as a problem in the early stages, when experimentation is limited to a few different data sources, types and metrics. But, as the number of sources, types and metrics increases, things could get very confusing for the end users of analytics data (who often don’t want to work very hard at understanding reports). We should know more about how Google plans to handle this issue as they begin to release a series of new reports in UA.
Bottom line: UA is an exciting development that holds significant promise for solving some difficult issues such as multi-device measurement and online/offline integration. For now, what is desperately needed is broader experimentation in order to test UA’s promise in real-world situations.
So, what things are you trying with UA? Please share!