Looking for Meaningful Data in Google Analytics
For the past three months I’ve been using the 4Q questionaire by Avinash Kaushik and iPerceptions but this morning I’ve removed it. The tool is a permission based survey tool that asks 4 questions:
Based on today’s visit, how would you rate your site experience overall? The answer is a 1 – 10 scale of satisfaction.
Which of the following best describes the primary purpose of your visit? Here they allow you to choose 4 options from about 165 possilbilites plus a fifth for other.
Were you able to complete the purpose of your visit today? Simple “yes” “no” answer.
What do you value most about the website? An open ended question.
Recently they’ve added a couple more options to include:
Which of the following best describes how often you visit this website?
How did you arrive at the website today?
These are great questions. A couple of these questions your analytics just won’t let you get at and a couple more which aren’t immediately apparent. So why am I removing it? At the moment this blog gets about 1,500 visitors per month. In the last month only four people agreed to take the survey but everyone had to read the introduction screen and make a choice about taking the survey. It’s a small barrier to entry, but in order to put up any barrier there needs to be a significant return for the Website owner.
I still think the questions Avinash is asking are important and having installed the survey has caused me to look at the data in my analytics and think about how I can get at that data from my current analytics provided by Google. Here’s what I’ve come up with.
Keywords. If you just look at your keywords for what terms your site is showing up for, your missing important information about why people visit your site. On this site, wp e-commmerce themes and variations on this theme accounted for 5 of the top 7 keyword phrases that the site was found for in the past month. Drilling down deeper 78 of the 520 keyword phrases the blog was found for were related to wp e-commerce themes. What I can see from the keywords is that these visitors have two primary purposes for their visit. Those who are looking for wp e-commerce themes, download wp e-commerce themes and the like are looking for a product. A much smaller number are looking for information about wp e-commerce themes with searches like “how to use wp ecommerce theme.”
In terms of task completion I can get some pretty good information around this from analytics. By adding event tracking to downloads I can get a good picture of the number of people who were looking for a theme and downloaded one. It’s fair to say that they were able to complete the task they came for.
In terms of the two recently added questions, Which of the following best describes how often you visit this website? And, How did you arrive at the website today? I can get to these via “Visitor Loyalty” under Google Analytics. For example I can see that 10.15% of my visitors have visited twice and .21% have visited between 51 and 100 times. And of course I can get at how they arrived on the site by looking at the “Traffic Sources” reports.
This leaves me with two questions that 4Q asks which are interesting and not immediately apparent through Google Analytics. How does the visitor rate their experience on the site overall and what do they value most about the site. Some people will say that you could look at the time on site, but someone can spend a long time on your site and be very unhappy with their experience. And there’s absolutely no way to get at what they value most, at least not through analytics.
For the moment I don’t have an alternative way to get this information, but certainly it’s something that I’ll be considering. I am also not suggesting that you not try 4Q on your site. I am saying that much of this information can be gotten at through the unobtrusive analytics you’re probably already using on your site.
Hi David,I agree with the sentiment of your post and reminded me of something troubling me regarding analysing visitor data and search engines. You mentioned visitors that spend a long time on a site but don’t have a good experience, suggesting that they can’t find what they are looking for.
The flip side of this is Pageviews per visitor or bounce rate. Perhaps a site is so well organised, SEO’ed and put together that every visitor that comes from search, finds exactly what they are looking for, converts and then leaves happy.
There has been talk of Google and the like using bounce rate data to determine the match between searcher intent and destination. I don’t see how these metrics could become a significant part of any search algorithm for the reasons above.
Perhaps the whole voting, bumping and reviewing democracy (which would seem to be too easily manipulated) will grow and the more obtrusive 4Q style questionaire will become a bigger part of our online experience in search.
Rob
Rob, I agree with you. I think the bounce rate and time on page are pretty easy to manipulate (depending on the traffic the site gets) and I don’t think that it would make much sense for Google to make these a significant part of their ranking algo. The thing to remember though is that they have over 100 ranking factors. Maybe 10% of these play a major role in how a site ranks the others play a minor role. This means two things, they can afford a little slop in those and in really competitive fields if you’re trying for a first place placement in the SERPS the minor factors will be what give you the extra boost.