Adding Meaning to Metrics on Blogs, Plus Commentary on Google Analytics' Intelligence

I recently read/watched a blog by web analytics evangelist, Avinash Kaushik, on the new Google Analytics Intelligence features. At this time, I haven't had a chance to play with them, since the feature is still being rolled out and my account does not yet have access, but I was inspired and wanted to discuss a little bit about putting meaning to metrics. In this post, I'll be focusing on blogs, but some of this could easily be adapted for use in e-commerce and news sites.

Anyone with any kind of website can get Google Analytics. It's free and easy - on the surface. You embed the script in your site's code and about 24 hours later, you start getting some information on your site's traffic. Notice the emphasis on the word traffic. Traffic is a metric in the sense that it measures something, but in order to really be web analytics, you need to put meaning to that traffic. In other words, you figure out what that something is that your traffic is measuring.

Say, for example, you're a blogger who has Google Analytics running, and you notice that your most recent blog post got 50% more visits than the average. That's a statistically significant increase in raw traffic, but what caused that blog post to get more traffic? Where did they come from? How did they find it? How engaged were the visitors?

Given the above questions, I'm going to go out on a limb and say that in the grand scheme of things, raw traffic means absolutely nothing. The fact that someone clicked on a link does not translate to quality, and unless you're getting paid per impression, you do not benefit from an increase in raw traffic.

Think of it this way: On a blog, you're selling content. On an e-commerce site, you're selling products. Does an e-commerce site benefit from someone clicking on a product page and doing nothing else, or worse, hitting the back button or closing the browser? You might see a slight increase in product awareness from that, but even that relationship is tenuous.

So, if raw traffic is meaningless and we need to measure engagement, what metrics should we look at? While not all-inclusive (I will keep this brief since this post is already long), the two things you'll want to measure are:

- Are visitors staying on the page long enough to actually read the content?
- Do I have a low bounce rate? (Generally, bounce rate measures how frequently people leave your site and go to another site.)

So why is measuring visitor engagement on a blog important? Well, for one, it tells you what types of content work best for your audience, so you can create more of that and less of what they don't like. This helps you keep the audience you have, and ideally, gain a larger audience (perhaps even through word of mouth) because you have a consistently good product. Additionally, and perhaps more importantly, if your site is monetized through affiliate advertising, either cost per action or cost per impression, more time spent on the page means more exposure to the advertising on your site.

Once the new Google Analytics Intelligence features roll out, you'll be able to set up custom alerts for these metrics, so you can create segments for your best and least performing content without having to dig. For example, you could set up an alert for content that has that tragic 90% bounce rate, but you could also set up alerts for content that has a magical 10% bounce rate.

I would go into further detail, but I don't have access to the tool yet, and besides, Kaushik does a much better job of explaining Google Analytics Intelligence on his site. Make sure you check out the videos, they're the best part.