Comparing Adobe Analytics to Google Analytics
I’ve now been a web analytics consultant for nearly 6 years. I’ve primarily focused on Adobe Analytics (formerly Omniture SiteCatalyst/Discover/Insight) and Google Analytics. Many of my clients that implement Adobe Analytics opt to also implement Google Analytics as well. Partially because it’s free, partially because they think comparing the data sets from two different tools may give them some additional insights, and possibly also give an indication if one tool is having tracking issues. I’m going to make my comparisons based on two of the top areas that I’ve had to handhold my clients through when transitioning between these two tools – custom tracking and campaign tracking.
Custom Variable & Event Tracking
Google Analytics does allow for some custom variables (maximum of 5), however they are somewhat limited in comparison to their Adobe Analytics counterparts and without the option of persistence (more about that later). In the new Universal Analytics, custom variables have been replaced by “custom dimensions & metrics.” The big difference now is that custom dimensions are handled server side, while the older custom variables were handled client side. They now allow a maximum of 20 custom dimensions. Custom events and campaign tracking with Google is also restricted by several factors. First of all custom events are limited to the following components: category, action, label, value and implicit count. If your business has a requirement to track events more granularly than that, then you’re out of luck with Google.
Adobe I think is the clear champion when it comes to custom tracking. The amount of custom variables (divided into two different types – traffic and conversion), not to mention event tracking and campaign tracking. Each of the two types of custom variables in Adobe Analytics allows up to 75 variables. The traffic variables are not persistent and are good for tracking things like internal search keywords, and any thing else you can think of that you wouldn’t want to persist from one page to the next. Conversion variables do persist, and you can control for how long (right done to per page view if you like, or per visit, per event action, a defined time period, or even never). As for event tracking, Adobe Analytics allows you to set up to 100 success events, with the ability to control if it is a counter (simple increments), currency or numeric – all with or without subrelations. These events, depending on how you set them up, can later be applied as metrics to your custom variables allowing for some powerful, in-depth analysis.
Custom event tracking is also a lot more powerful in Adobe Analytics, which allows you up to 100 custom events, in addition to some handy predefined success events for eCommerce tracking – including the ability to track revenue, orders, units, cart opens, cart views, checkouts, cart additions, cart removals and product views.
Google Analytics does a decent job with campaign tracking, but it’s extremely limited in what their query string parameters can capture:
- Campaign Source (required): This is used to identify the source of the data (google, eetimes, etc.)
- Campaign Medium (required): This is used to identify the medium type (email, rss feed, search, etc.)
- Campaign Term (optional): This is only used for PPC campaigns, used for the keyword term for the campaign.
- Campaign Content (optional): This can be used for A/B testing, content-targeted ads, to uniquely identify ads or links that point to the same URL
- Campaign Name (optional): This is to identify the specific campaign that’s running, to differentiate from other campaigns that may be running in parallel
These are the only query string parameters that you can define for any Google Analytics campaign, and while the Adobe campaign classification system allows you to apply classifications to non-campaign variables, Google pretty much only allows this and that’s it. It’s clean and simple, but a bit inflexible.
Once again I have to side with Adobe Analytics when it comes to campaign tracking, because of the greater flexibility and options available. You can customize your query string parameter(s), and you can use as many as you like (or as few, even just one). Then within Adobe Analytics you can classify the tracking code you used with your query string parameter to gain more meaningful insights into the data being captured.
Here is an example of how to classify your tracking codes. The “key” is the tracking code (notice how completely obscure they can be, if you want to keep them simple), and then how you can classify them – in this example by campaign name, medium, campaign start and end dates and “PMM,” which in this case means the “Product Marketing Manager.” You have the freedom to name your own classifications – whatever makes sense for your business, and you can also have many classifications (generally unlimited, but it is recommended to keep them under 50).
As you can see, this will give you a much greater degree of flexibility, and help you gain insights that are relevant to your business.
The other nice feature is the fact that classifying your tracking codes is completely retroactive. As long as you are capturing data for the tracking codes, you can go back and add, remove, and modify your classifications, and it will change them not only from the present going forward, but also change them going back all the way from when the data collection first started. Naturally that’s something to also be wary of – you need to consider carefully if a modification you want to make now, is one that you also want to apply to the past.
I know I sound very biased against Google Analytics. I do think it’s a fine tool for what it does, and it makes a wonderful “gateway” tool for anyone wanting to get into the field of web analytics. Google Analytics is a lot easier to implement and you can hit the ground running without a lot of customization, which isn’t true for Adobe Analytics. In my mind, I compare Google Analytics with ordering a simple cup of coffee at a diner, and Adobe Analytics with going to Starbucks and ordering a skinny triple-shot, half-caf espresso, light on the foam. With the diner you get simple options: black, with or without sugar, and with or without cream. Maybe you get the choice between decaf and regular. That’s it – clean and simple. With Starbucks you get a wide myriad of choices, and while it can be overwhelming at first (and more expensive!), it really is nice to be able to occasionally order a tall, no-whip latte with 3 pumps of peppermint. At the end of the day I honestly believe you get what you pay for.