QuestionPro Blog

This is a guest post from Gary Angel, President of Semphonic, a web analytics company based in San Francisco:

A Match Made in Heaven

The use of online surveys has grown dramatically in the past few years. And why not? Easy to deploy, incredibly inexpensive, reaching a large and increasingly representative audience, and providing near real-time information, online surveys are a dream come true for today’s researchers and analysts.

And yet (there is always an ‘and yet’), have you ever considered how much online survey research cannot easily tell you?

Using survey research, you can easily identify the core demographics that visit a site. You can identify visit intent – what visitors wanted to accomplish when they came to the site. You can identify overall site satisfaction and brand impression. With some clever use of sampling and serving techniques, you can even determine brand impact – how big a difference the experience of visiting your site actually made. If you ask a large number of questions, you can even get a pretty good sense of how specific elements of your site – like the navigational structure or the search or the section devoted to topic X were rated by your visitors.

This is all good stuff. 

The Missing Link

But what’s missing in all of this, is an understanding of what these visitors actually did on your site – and how the attitudes, ratings and demographics captured in your survey were driven by or drove actual site experiences.

Can you tell, for instance, whether visitors who are more satisfied with your site are actually more likely to return? Or is returning to your site more strongly correlated with something else – demographics, intent or some form of actual site behavior?

Can you tell what visitors who expressed dissatisfaction with your site actually did? Was it any different than anyone else? And if it was different is it different because they did those things – or did they do those things because they started out different?

If a visitor expressed high-satisfaction with your search (or strong dissatisfaction), do you know why? Can you tell, from among the 50,000 different types of searches on your site which ones help create satisfaction and which ones help create the reverse?

Can you tell if visitors even did what they said they were doing?

This would all be pretty useful stuff to know.

What Behavioral Data is Lacking

You may be surprised to learn that those of us who deal primarily with behavioral data have almost perfectly symmetrical set of problems.

With a web analytics tool like Omniture’s SiteCatalyst, we can track the movements and precise behavior of every visitor to the web site.

We can see who looked at what and when. We can track who returns to the site and who doesn’t. We can tell who searched, what they searched on, how many results they found, what they clicked on, and what they ended up doing.

But we can’t tell the first thing about who they were. Male? Female? Young? Old? These aren’t revealed in the behavioral data.

And we can’t tell you whether visitors were satisfied or not.

Did they leave the site because they found what they were looking for? Or did they exit because they couldn’t find what they are looking for? From the actual behavior, it’s impossible to know.

We know a visitor made a purchase. But did they end up buying despite their frustrations? Did they walk out of our site less happy than when they went in? From the actual behavior it’s impossible to know. What we measure as a success using behavioral data might actually have been a significant failure.

So there is a lot that students of traditional survey research can learn from behavioral analysis. And there is just as much that students of behavioral analysis can learn from online survey research.

Wouldn’t it be nice if you could combine the two sets of data? If you could see how the respondents to any given question actually behaved on the web site? If you could see how the people who searched for Product X rated search vs. the people who searched for Service Y? If, in any research you were doing, you could see the behavior, the attitudes, and the demographics of the visitors involved?

Well, it’s actually not very hard to do. Integrating your QuestionPro online surveys with a web analytics tool can be done – and done with surprising ease. In the next post, I’m going to show you how easy it is to integrate a QuestionPro survey with Omniture’s SiteCatalyst – the leading tool in the enterprise web analytics space. I’ll walk you through a series of increasingly sophisticated levels of integration – culminating in the full integration of every survey response from every survey respondent directly into the web behavioral data on a real-time basis.
This combination of survey research and behavioral data is a powerful combination. Once they are joined and you have expanded your view of the visitor to encompass both survey and behavioral data, you will surely want no man to tear these two asunder!

See part II here.

More Info:

• Omniture’s SiteCatalyst: http://www.omniture.com
• Semphonic, a web analytics consultancy: http://www.semphonic.com/

[Gary Angel is President of Semphonic, the leading independent web analytics consultancy in the United States. Headquartered in the San Francisco Bay Area, with offices in Washington, D.C. and Boston, Semphonic works with all of the major web analytics tools including Omniture, WebTrends, Unica, Google Analytics and Coremetrics. Semphonic clients include companies like the American Express, Barclays, the BBC, Charles Schwab, Genentech, Intuit, Kohler, the National Cancer Institute, National Geographic, Nokia, and Turner Broadcasting.]