Google Analytics vs. Adobe
So many people have written about the pros and cons of Adobe versus Google Analytics (GA). A quick search on the comparison will bring in a huge number of opinions. As a tracking code solutions provider with a considerable interest in the debate, what fresh perspective can we offer? Let’s start with a bit of background.
Origin of GA–launched in November 2005
GA caused a huge wave throughout the industry. GA’s biggest disruptive element was that it was a free solution. A lot of people were inclined to dismiss it for just that reason — and some people still do. I teach an Analytics class at a local university, and I just hosted a guest lecturer who called GA a toy. He went on to say, “if you’re doing business work, why would you bring a toy into the office?”
I don’t agree with that. Just because it’s free doesn’t mean that it’s a low-quality tool. From day one, I have been impressed with GA — and it definitely sent the industry scrambling. From the outset, Adobe/Omniture dug in and tried to differentiate itself as the vendor for large enterprise organizations. In turn, GA focused its message around the fact that it is easy to implement, and that anybody can use GA on their site. The industry players in the middle just scattered, and most of them were acquired in the ensuing years. Fifteen years ago Webtrends was the leader in the field and is now limping along. Coremetrics was acquired by IBM, and while I still run into analysts who use it, I often learn about that fact during a conversation about their search for an alternative solution. Adobe is still thriving, by pitching the kind of message that our guest speaker was touting — ”this [GA] is a toy.”
Lay of the land, background, what the field looks like
According to folks I’ve spoken with at GA, 85% of all websites in the world are tracked with GA. By Adobe’s reckoning, eight out of every ten dollars spent online are tracked with Adobe analytics. To me what that says is that the landscape has largely been carved up along those initial battle lines. Companies doing big sales, and large consumer-focused businesses (banks, retailers, service providers, etc.) are mostly using Adobe instead of GA — though occasionally they’re using both. A typical large enterprise business is not currently using GA alone. If you’re a small website, on the other hand, you’re not spending thousands of dollars a month to use Adobe — that’s what it would take, and you don’t need to. You could say that the lines have been drawn, and that the the war of attrition has led to a standoff between the two major players.
A change in the wind?
Barring another major disruption, I expect the current situation to hold. However, as GA continues to improve its offering, with products like GA 360, I keep expecting Adobe to counter with moves that make their implementation easier. They haven’t done that yet. Right now, Adobe continues to emphasize its ability to be highly customizable. Not long ago, I read about Adobe extending its cache of custom variables from 100 to 1000, which radically expands the number of metrics they can send. When I saw that, I thought, man, watching those 100 customizable variables was already like herding a hundred cats through a yarn factory every day, and now you’re giving me 900 more?
I know one guy that wants to lure large enterprise analysts away from Adobe, by teaching them how to meet their analytics needs in GA. Adobe prides itself on being customizable, and GA is more of a cookie cutter solution, but as the analytics world grows in complexity that’s not always a bad thing.
The fact that GA is less customizable means there’s less that can go wrong. Unfortunately, if something does go wrong and a piece of your data puzzle goes shooting off, you’re not always sure where to try to reattach it. With Adobe’s emphasis on widespread customization, there’s more that can go wrong, but data repair is easier — customization is Adobe’s strength, as well as its weakness. The tension between data customization and the risk of error is a gap that doesn’t go away. So far.