Features to Look for in Web Site Analysis Tools

Web site traffic analysis can range from simply counting unique visitors, to predicting what they'll do based on neural agents. To understand better what your company needs, it helps to know what is available and how it all works together.

Here are typical features you should look for in a web analytics package:

• Statistical Overlay: All analytics software records page hits, session length and unique visitors. Look for software that takes it a step further and actually gives you this data integrated with the page itself. That way you can go to a link on your site and see data on the number of clicks on that link next to the link itself. This feature makes it a lot easier to visualize what your visitors are doing.

• Dashboards: Dashboards are executive tools that collect the various stats and put them in one place for quick reference. You should be able to define what analytics you want to track and how you want to see them. With flexible dashboards, key personnel in your organization will be able to see at a glance how the site is performing, from their chosen perspective as a marketing executive, for example, or as the content manager.

• Clickstream Analysis: A basic feature of any analytics solution, clickstream analysis, also known as path or navigation analysis, tells you how users travel through your site. At its most basic, this looks like source/destination page referrals. A more robust clickstream analysis solution will use branching diagrams to show you precisely what route visitors are taking from beginning to end. But beware: This type of data can quickly become overwhelming unless you have a strategy for analyzing it.

• Custom definitions: Look for an analytics solution that builds on clickstream information by allowing you to define the behavior you want to track. For example, if you defined "conversion funnels" (see below), you could compare the actual behavior of your visitors with the specific paths you want them to take. Such an analysis will tell you whether your users are doing what you want them to, things like following a specific series of clicks to make a purchase, sign up for a newsletter, or reading several articles instead of just one per visit.

• Market Segmentation: Even if you're not in marketing, you know that customer populations can be sliced and diced any number of ways. An analytics solution that offers market segmentation lets you look at usage data for specific populations, such as a demographic or psychographic segment, as well as by site usage. For example, frequent visitors may use your site one way while infrequent or one-time visitors use it another. Or visitors who come to your front page through an opt-in email take one path through the site, while those who click on a banner appearing in the right margin of xyz.com take another route.

• Campaign Tracking: To maximize the return on investment for your marketing initiatives, you need to know how much traffic they’re generating, the number of orders, the value of the orders, and how the resulting revenue compares to the cost of the marketing campaign. Your analytics solution should help you track keyword buys, e-mail campaigns, banner ads, and other forms of marketing, from clickthrough to goal. With the right reporting tools, you can drill down to a dollar value for individual marketing elements or compare price/volume movement to help you devise new, more cost-effective promotional strategies.

• Conversion Funnel Monitoring: Suppose your customer has a 6-step process from banner to buy:

1. Click on an ad banner --> View product page
2. Click “Add to Shopping Cart” --> View shopping cart
3. Click “Check out” --> View order form
4. Fill out billing/shipping info- -> View payment options
5. Fill out credit card info --> View order confirmation
6. Confirm order --> View Thank You

Assuming the visitor makes it to the first step and lands on your site, there are still five more opportunities to lose the sale. Fewer and fewer people make it through the process at each step. When charted, this pattern looks like a funnel that appears narrower or wider, depending on your conversion rate. Whatever it looks like, a good analytics package will tell you exactly how many people are dropping off at each stage. This gives you the clues you need to identify and remove barriers and increase conversion.

• Form Abandonment: You'd be surprised how much damage an overly ambitious form can do to a site. An analytics package that tracks form abandonment will tell you the field on your form that caused a visitor to give up. For example, as the manager for a lead-generation site, you may see a contact request form as the perfect opportunity to gather geographic information. But if asking for zip code is causing 40 percent of your visitors to walk away, then you have to ask yourself if that extra bit of information is really worth it. An analytics package with this feature will give you the data you need to make that decision.