Page views, distinct hosts, conversion rates, and visitor segmentation. It’s enough to make any small to mid-size business owner’s eyes glaze over. However, those are but a few of the myriad of terms that crop up in the science of Web analytics.

“There are literally hundreds of performance indicators. There’s so much data available it’s easy to get overwhelmed,” says Megan Burns, a senior analyst from Forrester Research, the Cambridge, Mass. IT research firm.

It’s also easy to get confused with what Web analytics are for in the first place. “Most business owners don’t pay enough attention to Web analytics. Usually, it’s because it’s harder than they expected,” says Eric Peterson, author of Web Analytics Demystified.

Reasons to use Web analytics

Generating raw data is what Web analytic tools do best. This data is ultimately about the behavior of your online customers, what pages or features they visit on your website, and ultimately whether they buy your products. When the numbers are added up, you can get metrics that can tell you a tremendous amount about whether your Web strategy is working.

There are a variety of Web analytics tools available today, from free products to services that will help analyze your results. Depending upon which analytic tools you deploy, you can find out an array of information about your customers and therefore your website, such as:

  • What is the geographic location of your customers?
  • How long do customers spend on certain pages?
  • Which website features generate the most hits?
  • Which landing pages most often lead to customer purchases?

Challenges of Web analytics

All that data is great to gather, but before a business takes the plunge into the world of Web analytics, you need to ask yourselves some key questions about whether your business has the time, money, and staff to put the data to good use. Interpreting Web data is not a job for the webmaster or the IT person who keeps the network up and running. At larger companies, there are entire Web analytic teams. Most small businesses cannot even afford a full or part-time Web analyst.

More likely the boss, the finance director, or marketing director will have to get up to speed in interpreting Web data. “The dirty little secret is that Web analytics is hard. Companies say it’s easy. The truth is its difficult to learn and requires effort,” says Peterson.

The good news is there is a very active community of Web analytic practitioners. The Web Analytics Association is a good place to start. It offers articles, as well as links to other resources, training and workshops.

How to make Web analytics work

Interpreting the data and determining it’s relevance to a specific business is where the challenges begin. Here’s a roadmap of how to get the most out of Web analytics:

  1. Define the goals of the website itself. What makes the site successful for the business: generating leads, selling products and services, or driving lots of traffic to serve advertisements?
  2. Target the performance indicators that will define success.For example, in the case of an e-commerce site, the relevant data may be the number of orders generated or the “add to cart” conversion rate (the ratio of people who added a product to their cart, but didn’t finish the transaction). On the other hand, a site that makes its money from serving ads would be looking more at impressions (the raw number of advertisements downloaded within a page), average number of pages viewed per visitor, or the frequency of their returning to the site.
  3. Monitor those indicators on a regular basis.Whether it’s once per day, per week, or per month, it’s important to track those benchmarks on a regular basis. The numbers need to identify trends, not flukes in visitor behavior. Consistent vigilance is the only way to distinguish between the two.
  4. Diagnose the results properly.It’s very easy to misinterpret data. For example, one might assume a visitor’s time spent on the site was time focused on the content. In fact, it may just mean someone clicked onto a page and left it open while checking their e-mail in another window. “Think of Web analytics as part of a toolkit of techniques and technologies to diagnose the health of your website," Burns says. "It measures behaviors like what visitors do, what they look at and where they abandon the process. What it doesn’t tell you is why.”

Free solutions or paid?

There are a number of free tools available. Google’s analytic tools are, by far, the most popular to provide you with basic insight into how people find your website and determine whether to buy keyword-based advertising on Google's search service. However there are paid solutions that are sometimes worth the cost, as well. Peterson highlights the differences, listing the typical upgrades that come with a price:

  • Visitor segmentation behavior: This gives you the ability to drill down and compare the activities of different audiences. This may be a comparison of visitors from different geographic locations, new versus returning customers, or customers who buy from one product line versus another. Such comparisons can offer a treasure trove of insights for marketing managers.
  • Importing other data. If a company can mash up accounting information with Web performance indicators, then calculating the gross margin per visit becomes possible. In addition to measuring the health of the site, now it’s helping to measure the health of the business itself.
  • Tech support and education.What good is a tool if it’s not being used to its full potential? “The best way to learn about Web analytics is to have someone to call who can answer your questions,” says Peterson.


One of the most interesting aspects of the digital age is the amount of data that is generated about consumer behavior -- and therefore about business. If your business has the resources, Web analytic tools can provide insights that it would take your business years of analyzing sales reports to comprehend. The free tools are fine if you have the expertise to make use of the data. Peterson advises that it's sometimes worth it to pay someone else to help interpret the numbers. Businesses should plan on spending around $7,000 dollars up front for software tools and $500 to a $1,000 a month, thereafter, in service fees. That may sound like a major expense, but it can be pennies on the dollar if companies can help translate that information into having a positive effect on the bottom line.

SIDEBAR: Web Analytics Vendors

Some of the paid solutions on the market today for small to mid-size business owners include:

ClickTracks -- One of the least expensive paid solutions on the market. The visual interface helps answer questions about why your website visitors do what they do, how your search engine optimization plan is working, and whether your website is earning ROI.

Visual Sciences – This company provides its clients, regardless of their analytics expertise, with applications to better understand and optimize website and Internet marketing programs. Visual Sciences offers real-time reporting and tailored applications for your business.

Coremetrics -- Online Analytics product captures every click of every visitor over time so that they can give you the intelligence you need. Features include a 3D view of the effectiveness of your marketing campaign, insight into shopping cart abandonment, and content analysis.

WebTrends – This analytic product not only gives you data about customers but enables you to see if your Web 2.0 investments are paying off by measuring and testing visitor interaction with AJAX, online video and more. It also helps you identify content, channels and campaigns that best keep visitors engaged.