Sublime

Think of creative ways to engage traffic—what can you do to keep a Visitor for 60 seconds or more? Segment the Visitors who stay more than two minutes, and analyze the content they consume, where they come from, and so forth, so you can learn what you should be doing more.

My friends Shane Atchison and Jason Burby are authors of the book Actionable Web Analytics. They have long advocated creating models that identify the cost of delays. Figure 14-7 shows one such model.

You are a part-time analyst or a professional consultant. You are hired to do Omniture analysis at a company, and you find that even a 3–5 percent error is a big deal because of small overall numbers for the site. You then spend hours and days trying to make data collection more accurate. This is a terrible use of time and resources. If you work

in-depth analysis of customer churn (meaning identifying those who recently abandoned the product)

If you are a small site, focus on the actionable head data.

For this reason, I recommend a focus on customer experience and on experimentation and testing. To truly exploit these approaches for your business, you need to be good at statistics.