Site statistics play a critical role in my online business. But you might be surprised by the sorts of things that I pay attention to.

Most new Internet professionals focus on a few data points such as hits, pageviews or unique visitors. While those data points are fine, they are what I refer to as "passive stats" - they don't give you much in terms of actionable steps that you can take to improve your site's success. Plus, passive stats are too easy to manipulate and artificially inflate.

Instead of focusing on passive stats, you should focus on active stats. Here's how.

1. Monitor traffic sources

You should constantly be looking at your referral logs to see which sites are driving traffic to your site. As you experiment with various marketing and promotion techniques, you can quickly learn which are working and which aren't. Which blogs are worth commenting at? Well, the ones that actually send good traffic to your site. As you learn the types of marketing actions that result in targeted traffic, you can optimize the time you spend marketing.

Read the rest of this entry

Though people might be interested in using this:

<a href="http://pmetrics.performancing.com/6"><img src="http://performancing.com/images/125-125.gif" alt="blog statistics" />

Just make sure to replace the "6" with your own affiliate code.

I am not really a pro blogger and I am not earning my living from blogging. Nevertheless I use AdSense and Text-Link-Ads mainly in order to maintain the motivation level. Today something unusual happened. AdSense notifier told me that suddenly one of my clicks generated 3 times more clicks, than usually. Naturally I was interested to discover what happened: some post jumped to the first page of Google search results or maybe some A-blogger referred to some post of mine.

What puzzles me is that neither Google Analytics, nor SiteMeter, nor PMetrics don't have an easy solution for me. What I often want from the stat package is to discover where some particular change comes from. Unfortunately most if not all the packages are aimed at detailing the static picture, not the dynamic one.

What do you do when you need to solve similar problems? Do you download the stats data to Excel and examine the changes in it? Or is there some special report in the Google Analytics that could be helpful?

For those of you who have been following the new soap opera "pMetrics", let me put things in context. The management structure of Performancing did not have enough built-in redundancy, and with David Krug's sudden and surprising extended absence, there just wasn't an emergency plan in place, so pMetrics suffered as a result.

That's not an excuse. We accept responsibility for the mistake and are taking the steps to ensure that Performancing operates as a redundant, decentralized team so that its operation doesn't depend on any single person.

While those steps are in motion, we are happy to be able to announce that pMetrics will be back in operating condition this week. The getclicky team is phenomenal and we are superpsychked (new word?) to be hooking up with them on this venture. We hope you'll stick with us and enjoy the fruits of this collaboration.

On a related note...

Is there any interest from the Performancing community in having the old pMetrics software opened up for community development? Any thoughts?