An Analysis of My Data

I’ll be honest, it has been a long time since I looked at the analytics for this blog, I just haven’t had any time. Work has been flat-out busy since January 1. That’s also the reason that my posting has been very light.

Anyway, I logged into GA today to check some stats and was surprised by some of the data. I thought you all might be interested in the basic process I used to figure out what happened.

The Big Picture: What’s happened Since January 1

Since I haven’t reviewed the traffic data for a couple of months I started by adjusting the date filter to cover January 1 to March 4. Here’s the Executive Overview report:

Notice anything? What the heck happened around February 1st? Obviously some website drove a ton of traffic to the blog. Was it Digg?

Drilling Down: When did it Happen?

Before I get ahead of myself and try to figure out where the traffic came from, let’s determine exactly when it happened. I’m going to use the date filter to drill down and isolate when the date happened. If I hover my pointer over the data above GA shows that the spike occurred on February 1.

If I adjust the date filter one more time to cover February 1 we can see there was a big jump in traffic at 8 AM:

Now we’re getting somewhere. Now that I know when the spike happened I can start to figure out where the traffic came form?

Getting Closer: Where did they come from?

I’m trying to figure out where the traffic came from, so I’m going to look at referral information using the Marketing Optimization > Visitor Segment Performance > Referring Source report. This report segments the site traffic based on where it originated. When we look at this report we should immediately know where the traffic came from:

A ha! I was Stumble-Uponed! For those of you unfamiliar with StumbleUpon, here’s a description from their website:

People-Driven Technology:
Using a combination of human opinions and machine learning to immediately deliver relevant content, StumbleUpon presents only web sites which have been suggested by other like-minded Stumblers. Each time the ‘Stumble’ button is clicked, the user is presented with a high quality web site based on the collective opinions of other like-minded web surfers.

Based on this description I’m going to say that StumbleUpon drives qualified traffic to my site. We like qualified traffic, it usually converts better :)

Is there anything else to know?

What else can I learn about these people? First, they didn’t convert on any of my GA goals! I can see this in the above report. There is a 0% conversion rate for Goal 1 and Goal 2. This means that they didn’t do what I wanted them to do. (The goals for my website are to sign up for the RSS feed and to use the contact form). What’s disturbing is that this was qualified traffic! Is this indicative of a problem with my website? Why didn’t these people convert?

I’m going to dig deeper by cross segmenting the data in the Referring Source report. I’m trying to find out more about these people that came to my site from StumbleUpon. Did they come to the site before (i.e. how many return visits?):

Not too many return visits, almost all new visits. So I now know that StumbleUpon drove 136 new visits to my blog in about one hour and none of them converted. Wow, that’s a complete bummer.

Summing Up

I could dig a bit deeper, and look at how the visitors from StumbleUpon navigated the site. But, with an average number of pageviews below 2 I’m not going to discover too much. There are some lessons to be learned here:

First, I should have been paying closer attention to my reports. I probably don’t need to review them daily, but I should review them once a week. Also, I’m going to dig a bit deeper into my new visitor segment. I really think I may have a problem converting new visitors. But that’s another post for a later time :)