Posted to Ben Finklea's blog on October 6th, 2010

Six Common Google Analytics FAILs

Volacci’s marketing coordinators have been sharpening their Google Analytics wisdom and wit over the last year or so, and have Google Analytics certification to prove it. We’ve been analyzing stats, crunching numbers, running audits, and creating marketing strategies from our insights. Why do I bring this up, you ask? 
It’s that crucial time again when most brands and businesses prepare for their fall and holiday (e-commerce) campaigns. While consumers may not be thinking about it yet, just ask Pottery Barn when they started stocking Christmas decorations! When preparation begins, one of the first things you should be thinking about is your web analytics and correctness of data, especially in an economy that is slowly recovering. This winter could potentially be a huge corner for all of us to turn. If your analytics are off by a “smidge”, or even by a nudge, you’re whole fall and holiday roadmap may for inaccurately assessing current baselines and what is considered to be a “successful” benchmark to hit. Even worse, the wrong data can cause your team to draw inaccurate conclusions, such as classifying paid search traffic and conversions as produced organically. If you can’t accurately tie results to their source, you’re bound to make some utter mistakes. Or, as our online culture likes to call it: FAIL So, before you go bounding off to an important meeting to make an misinformed decision on your fall and holiday campaigns, please learn from others’ mistakes and never commit the following six common Google Analytics FAILs. #1: Testing FAIL The only way you can know if something is going to succeed (or fail) is if you test it. If you fail to test, you will fail more than you succeed. With that said, web analytics should be used to solve problems and to make the online experience easier, faster, more compelling. Anyone in the e-commerce game knows that their greatest hunches and marketing instincts should be tested and quantified to what improvements they can provide. 
By using your analytics as a baseline, and running Google Website Optimizer testing tool, or the A/B testing features in AdWords, you can determine if your hunches and instincts are correct or not. #2: Tagging Traffic Source FAIL When you start your paid search campaign in Google’s AdWords, you get different tracking codes for traffic and for conversions. It is important that you differentiate these two sets of coding and install them on the appropriate pages. Conversions coding should be placed on final conversion page, where the final conversion is finished. Google also allows you to employ “autotagging” to effortlessly track whether your Google search traffic (and sales) is coming from paid or organic. Paid campaigns at Bing, Yahoo, and other search engines that don’t rhyme with “Boogle” will appear to be organic traffic unless you use campaign tracking links to tell your analytics otherwise. #3: Tracking Email FAIL This FAIL is incredibly important for e-commerce sites with email campaigns. Your main house email is the biggest-yielding channel. If you don’t use campaign-tracking codes, you’ll have no simple way to track the total revenue produced by your house email, or your overall conversion rate. You should also track other emails that aren’t directly tied to a specific email campaign, such as your abandoned cart email and other transactional emails. #4: Filtering Internal Traffic and Test Orders FAIL If you fail to filter out internal traffic and test orders, your analytics will feel a little too good to be true. In fact, everything from conversion rate to revenue can be completely inaccurate if your team fails to set up an exclude filter on test orders. You can still maintain an “all traffic” profile (that way you can see the outcome of the test orders), but have at least one profile that filters out internal IP addresses so that you know what the real numbers are. #5: Tracking On-Site Search FAIL As our society becomes more search-savvy, they will begin to use it in more sophisticated instances. And while using on-site search may not solidify anyone a country club membership, if tracked correctly, it can provide insight into the consumers’ minds. Not all web platforms are engineered to display the search term in the URL of the results page, but most are getting there. Google Analytics provides a basic reporting of top search terms, the overall share of users who use on-site search, and how much business they do. #6: Shopping Cart FAIL Get out your conversion funnel... and flip it upside down. Your shopping cart should be the first page of your conversion funnel, NOT your checkout page. The later you start your funnel, the more falsely inflated your conversion rate number will be. The cart is a critical page to track because it tells you your engagement rate as well. (this figure includes the percentage of site visitors who add anything to the cart) And it tracks the performance of the first page of the checkout, which is one of the most difficult transitions in e-commerce. Thank You For Reading! No one likes people who don’t share, especially giant flying cats. So if you liked what you read, please share my post with any of our socially-labeled buttons, or we’ll sick Fluffy after you! Please subscribe to our RSS feed as well so you can receive daily fodder from our blog.