Posted by Andrew Edwards on June 6th, 2011 at 4:13 pm

Is web analytics giving you a case of "agita" because it's complicated and not very pretty?

Rest easy. All you need to do is drop a line of Google Analytics code into the header or footer of your page and then...that's it! Analytics becomes beautiful and fun. It's that simple! Is that not amazing?

What's amazing of course, is how many people actually seem to believe that.

These days, no one actually wants to pay for an analytics product, and who can blame them? Except for large enterprises and those who pay special attention to the protection of PI, paying Adobe or WebTrends for their robust, heavy-load solutions makes less and less sense.

But what makes even less sense than paying for a vendor's toolkit is believing that because Google Analytics is free, then everything associated with it must also be in a way magical.

Like setting it up properly--that must be a breeze! No. Actually it is just as hard as setting up Omniture.

Like figuring out what you want to measure--Google does that for you! No. Actually it does not. You still have to bring in your analytics experts and your stakeholders and figure out how to turn your requirements into reports.

What about all that arduous tagging you needed to do before? Google is free, and that is painless. Therefore tagging for Google is also painless? Sorry. An expert still needs to write them, and QA them before they are deployed.

But Google can't ever be bested by another solution, can it? How can you beat "free"? Yet in fact, it can be bested. Many complex operations needed by power users of analytics just don't get done in GA. Doesn't matter how little you paid! Still won't do it. Big enterprises also want to own their data. They don't want Google to own it, so they refrain from letting Google analyze their web data. This means they pay for a solution--WebTrends, CoreMetrics, Unica...the list goes on.

Finally, since Google is free, doesn't that mean you don't need to plan how to roll it out to your various constituencies? And you don't need to pay anyone to configure it? Or to make sure it's accurate?

No, No, and no.

So there are two ways to move to Google Analytics.

One way is to believe that Google is both free AND easy. And that you don't need to have real analytics experts on the team to help you plan, or write custom tags, or make sure your proper analytics governance plan is in place, or make sure your marketers are not wildcatting their own little instances of Google Analytics to the benefit of exactly no one; and that you don't have to bother with any of the other prudent and success-focused behaviors associated with using other analytics tools--in fact, that you don't need to do anything at all except sign up for GA and drop a tag.

The other way is to think of GA as the very same tool you had before, except now you are not paying for the tool itself. In this scenario, you are still way ahead of the game. Because now you can take those same dollars and devote them towards very careful report planning, detailed tag-creation and QA, and a good deal of close monitoring of your entire analytics effort. After you've spent that money, are you farther along than you would have been with the same dollars spent just on obtaining a license for the tool itself? In most cases, absolutely! But making sure you don't just put the saved money away in a sock-drawer is just as important as the choice of Google Analytics in the first place.

Success requires that you leverage GA. Take the money you saved and devote it making a great GA reporting structure.

We've already said that one of the above methods succeeds and the other does not.

Which one most closely resembles your company's move to GA?