Learning to Use Google Analytics, Part 4

February 16, 2006 Dennis D. McDonald

One thing you can do with Google Analytics is compare the references your web site gets from different sources. This installment in my Learning Google Analytics series compares one weeks' visits referred from Google itself (e.g., from searches on Google that resulted in someone linking to my web site All Kind Food) with one one-week's visits that were referred from one particular web site where I had left comments and a link back here on a blog entry.

Google consistently shows up in Google Analytics as All Kind Food's most frequent referrer, despite the frequent (and I do mean frequent) visits by MSN's indexing 'bots and crawlers.

Google searches that come to All Kind Food are quite varied. Some are out and out errors and are caused by my last name (McDonald) being very similar to a certain large fast food company's name. For example, I've noticed that quite a few people when including MCDONALD and FOOD in the same search query arrive at my web site. I'm not complaining, that's just a fact, and I hope these visitors enjoy their stays.

People who arrive at my web site also are frequently are searching for book and movie reviews, and I'm quite gratified by that. For example, I've published a few reviews of Japanese and Indian films, and when I track back to the search query that arrives at my site I frequently find my review entry listed on the first page of the search.

Lets face it, though, readers of my book and movie reviews are probably not looking for management consultants to hire. In that regard, I'm especially interested in generating traffic to my web site's Managing Technology section. This is where I publish most of my my industry- and professionally-oriented items. (I also publish on The Podcast Roundtable web site.)

I have been noticing that direct referrals from other web sites or web pages seem to be more successful than search queries in driving traffic to Managing Technology. That's why I decided to make some comparisons between general Google referrals and more targeted referrals from web sites that are known (by me) to publish information related to my own consulting interests.

This past week I published comments and links on an article published on the Freedom to Tinker blog. The article links to a 27-page downloadable report that I reviewed in an entry in All Kind Food. After I published my review I left a note on Freedom to Tinker tht contained a link back to the review.

I immediately began noticing direct links from Freedom to Tinker to my review in my search logs. That led me to wonder, how do these links compare with the more varied links to All Kind Food that are generated by Google searches?

First I compared the referrals in terms of the browser that was used by the visitor. Here is a pie chart from Google Analytics that shows the relative distribution of browser types for visits to All Kind Food made by Google searchers:

Web Browsers Used by Google Referrals

About four times as many Google visitors are counted from the above chart as for the chart below.

Here is the distribution of web browsers used by visitors who linked directly from the Freedom to Tinker blog to my review here:

Web Browsers Used by Freedom to Tinker Referrals

For some reason, even though Microsoft Internet Explorer users are the proportionally largest group of referrals from Google, "Freedom to Tinker" direct referrals are more likely to be Firefox users. Are there any other differences between the two groups? Let's see.

Source Country of Google Referrals

One way to compare visitors is in terms of their source country. Here is the breakdown for Google referrals which shows that nearly 60% of the past week's Google referrals have been from the United States:

Source Country of Freedom to Tinker Referrals

Here is a breakdown of Freedom to Tinker referrals:

So it appears that, at least proportionally, "freedom to Tinker" referrals are more likely than "general" Google referrals (79% vs. 58%) to be from the United States. Is this perhaps due to Americans just being more "litigious" than others? Or is it because Freedom to Tinker usually deals with issues that are of particular importance to Americans? I can't really say but it is a difference worth thinking about, since I don't think that the issue of DRM (digital rights management or digital restrictions management - take your pick) is a peculiarly Americon-worrying issue. (Another simple possibility is that it's just impossible to answer these types of questions using this type of data!)

Google Referral Visitors: New vs. Returning

Another way to compare groups is whether or not the visitor is "new" to All Kind Food, i.e., has the visitor visited the web site before? Here's a chart that shows the "new vs. returning" breakdown for general Google referrals:

"Freedom to Tinker" Referral Visitors: New vs. Returning

And here's a chart showing the percent of "Freedom to Tinker" referrals that are new:

It looks as if about a third of the visitors to All Kind Food, whether they come via a Google search or a tightly-focused link from a referring site (that I planted myself), are returning visitors, i.e., they've "...been here before."

Frankly, I'm surprised by this and I'll be doing some more thinking -- and research -- to get a better handle on the explanation. Is this just a statistical anomaly, or is there an explanation that might prove useful as I seek ways to attract new readers? I can't tell yet.

If any readers can suggest some possible explanations of this close proportion of "new" and "returning" visitors, please email me.

____________________________________________

Other entries in this series: