Everything You Need To Know About Universal Analytics
But don’t slump in your chair and leave unenthused about Universal Analytics. It is really cool, but there’s not exactly anything to see in Google Analytics just yet. That doesn’t mean you should just sit on your hands until there is, however. Now is the time to plan.
You Need A Universal Analytics Strategy
Brainstorm your questions
If technology were no barrier at all, what would you want to know about how your users/visitors/customers interact with you? Ask yourself questions such as…
- Who are our audiences? (If you’re an ecommerce site) buyers and potential buyers? (If you’re a non-profit) members, high-value funders, press, volunteers? (If you’re an app or community site) subscribers, potential subscribers, troubleshooters? (If you’re a lead generation site) potential leads, certified prospects, existing customers?
- What would you want to know about each of those audiences to influence their experience? For example, what are the behaviors of a potential buyer/subscriber/member/lead that might indicate higher likelihood to buy/subscribe/donate/request more information? What do I want to know about marketing or communication with those audiences to influence their decision process? What can I do to make no-fun activities like troubleshooting and support as easy and seamless as possible?
Inventory your data sources
Your website is just one place your users/visitors/customers interact with you. There are lots of other places where you may already be collecting data, or where you have the potential to collect data, about those interactions that may help you inform answers to the questions you just brainstormed. Figure out what those are. Suggested places to look:
- Website(s) for online interactions
- Mobile app for interactions on mobile devices
- Marketing systems data — email marketing tools, survey tools, display and paid search advertising tools, etc. for interactions with your marketing campaigns outside your website
- Social network data about who likes/follows/comments/tweets/etc.
- Call center data for phone interactions in sales or support roles
- CRM, customer database, or point-of-sale system data for a whole-customer view, especially offline interactions like direct mail, in-store purchases, etc.
Think about your customer ID
Google Analytics used to just randomly generate a visitor ID that it stored in a cookie to identify returning visitors. You never saw this number, because it didn’t have any meaning except in that cookie. Universal Analytics lets you specify this customer ID to be anything you want. Most usefully, it could be an ID you can associate with the same person across many of those data sources above. A few pieces of advice:
- No PII! The terms of service (link goes to US terms, check your jurisdiction) of Google Analytics prevent you from collecting personally identifiable information into Google Analytics. You can record a numeric customer ID or something like that in Google Analytics and use that to associate with data from those other systems. Ideally, this is an existing unique identifier for all your customers/visitors/users. Do not use a name, email address, social security number (!), or anything like that.
- However, do think about how you can associate the same person with that customer ID across many systems. Ideally, you have a CRM or customer database or something where you can link many identifiers together: customer ID, name, email, phone number, credit card number, Facebook ID, etc. Then we can do things such as link website data (visitor ID) with call center data (phone number), email marketing information (email address), in-store purchases (credit card number), etc.
- Even aside from the terms of service for GA, make sure you respect customer privacy (including a clear privacy statement about how you use data and what you collect), give users the ability to opt out, and comply with all laws and regulations in the jurisdictions in which you operate.
Define your processes around data
Where do you want to be able to report on and see data? Will Google Analytics be your primary way of looking at data — you’ll import everything to GA and report and analyze data there? Or do you have an existing data warehouse and/or business intelligence reporting and analysis systems that you will use in conjunction with Google Analytics to give you the full picture?
- Getting data into Google Analytics is easy with the new Measurement Protocol. Website and app data are super-easy with pre-built libraries. Other systems may require customization, and you may need to stitch together identifying information about customers (see above) before sending the data to Google Analytics.
- Getting data out of Google Analytics offers many options that potentially allow you integration of data from multiple sources. Of course there are simple options, like custom reports and Excel or Google Spreadsheet export, as well as a number of tools (1, 2) that use the Google Analytics API to pull data out of Google Analytics. For Premium customers (who have lots of data), the new BigQuery integration will allow you to query your raw data and get the results.
Whither From Here?
Now that you have a better idea about what you want to measure and how you want to use the data, you are in a position to really think about how Universal Analytics can help you.
Universal Analytics is currently in beta. You have to create a new web property to try it out, which means no historical continuity with your old data. (Presumably, when Google decides it’s a little less “beta”, there will be a way to migrate your existing web properties to Universal Analytics.) But the good news for now is that you can run Universal Analytics (analytics.js) and legacy Google Analytics (ga.js) on the same websites without interfering with one another, so you can keep your existing data intact, but try out Universal Analytics. At the same time, you can begin sending data from other sources using the Measurement Protocol. (See one example in an earlier post on this blog, where Jim Gianoglio demonstrated bringing in email marketing system data such as open rates.)
As Universal Analytics continues to develop and additional features are unveiled, you can be sure we’ll be talking about them here to guide you through how to use them most effectively.
About Jonathan Weber
Jonathan Weber is the Data Evangelist at LunaMetrics. He spreads the principles of analytics through our training seminars all over the East coast. The next seminar he'll be leading will be a Google Analytics training in Boston. Before he caught the analytics bug, he worked in information architecture. He holds a Master’s degree from the University of Pittsburgh School of Information Sciences. Jonathan’s breadth of knowledge – from statistics to analysis to library science – is somewhat overwhelming.