Usability Testing – Beyond Google Analytics
Posted on February 25, 2011
The Internet offers businesses a unique marketing opportunity that was unheard of in the pre-digital age. We can, quite literally, monitor our customers or clients as they browse our website, learn about what we have to offer, and make purchase decisions. Old news, right?
Everyone uses Google Analytics, or some popular alternative. Even when many of the readers of this site were still children many websites used some sort of visitor monitoring, even if it was a simple hit counter displayed prominently on the page. Tracking visitors was simple then. We counted hits, and then progressed to pageviews. Now pageviews are a secondary metric. It’s all about unique visitors, conversions, and first touch attribution. This is all very difficult for a website owner to keep up with, short of hiring an Analytics expert. (Hint: If you’re running an online business, you absolutely should be paying an Analytics expert – or at the very least a Search Engine Optimization consultant who is familiar with Analytics.)
While basic Analytics packages such as Google’s offer a lot of insight into the customer’s interaction with the website, they still don’t tell the entire story. There is much more useful information to be garnered. Businesses of 20 years ago would have killed to get their hands on the kind of data that is available to us.
Watching the Visitor’s Page Interactions
It is true that Google Analytics offers some in-page data showing how your site’s visitors interact with your pages. They even improved it in a recent update. However, Analytics still lags behind some other tools in this area.
GET YOUR 100 WAYS TO MAKE $100K CHEAT SHEET PDF
(100% FREE DOWNLOAD CHEAT SHEET)
Consider this: you are a seller of high-end vacuum cleaners. These aren’t the kind you buy at Wal-Mart, but the ones that actually could suck the clothes right off your body and leave you standing in your boxers. Looking at Google Analytics you are able to see that visitors entering on your category and product pages tend to convert extremely well. If the first page they see is your home page, the visitor is half as likely to make a purchase.
Dig into the data all you want, it’s difficult to pinpoint exactly what needs to be improved. Your navigation is a little bit difficult on the homepage, but you are not entirely sure if that’s the problem. Some featured products are centered on the front area of the homepage, and even receive a substantial number of clicks. What’s the real problem?
Enter mouse tracking. Tools such as Crazy Egg and Click Heat can show you heat maps of where your users are actually clicking on the page. Not only that, they can indicate to you how long it takes the average user to find what they are looking for, and where they tend to hesitate. This data is crucial to nailing down a root cause of a problem. It also isn’t easily indicated by simply looking at the click data in Analytics.
Back to School – A/B Testing
You have identified a problem with the navigation on your homepage. It seems that visitors have trouble finding it and then tend to be a little confused on how to properly get where they need to go. While they seem to trust your brand – instead of immediately bouncing they stick around until they find what they are looking for – this slight frustration wears on them. Steve Krug might say that it saps away some of the Goodwill in their reservoir. Regardless, it needs fixing and you are ready to get it done.
After ponying up the money, your trusted web designer presents you with two great options for improving the homepage navigation. Which one should you choose? Why both of them, of course! How will you ever know which one is the best for your website’s users if you don’t try them both?
But wait! Putting one design up for a month and then replacing it with the other isn’t quite fair. Other forces change the attitude and mentality of your visitors in that time, so you can’t really be sure the design is the cause of any changes. You need to test them simultaneously.
Enter A/B and Multivariate Testing
If you are into the whole testing scene (which you better be if you run a commercial website), then Google Website Optimizer is really the best tool for the job. Others exist, such as Visual Website Optimizer. Once again, Google is free and very effective.
An A/B test simply allows you to put up two different variations of a web page and send your visitors equally to each one. After enough testing you will be able to see which version performs better. Multivariate testing just means you present more than two different possibilities. For most people, sticking with two is all you will need. In the case of our vacuum cleaner website, they were able to make an informed decision based on the data gathered during their simple A/B test. If you would like to learn more about testing strategies, you should look at as many articles as possible to make an informed decision.
Tying It All Together
As if being overloaded by the mounds of great data in Google Analytics wasn’t enough, now I have given you more tools to play around with. Not only that, but you’re going to start running tests. This isn’t getting any easier!
Maybe not, but these tools and techniques will help your website perform better. If you are like many other small businesses out there, you don’t have time to deal with heat maps and A/B testing. Talk with your web designer or your SEO consultant to see if they have experience in usability testing. If not, consider seeking a specialist.
The next step involves actually listening to your customers and receiving their feedbacks. Look for another post that details some tools and techniques to effectively utilize your greatest asset – the website visitor.