Web Analytics: Understanding Visitor Behavior

In the beginning, there were “hit counters”. Occasionally you will still see one at the bottom of a website, prominently displaying the number of hits over a given time frame. There was a time when the success of a website was gauged solely on this number, with a site that received many visitors being deemed “successful”. Boy, how times have changed.

Since the inception of the web, developers have sought to measure the success of websites through the use of web analytics, which is the process of collecting, collating, and analyzing a website’s activity. For ecommerce sites, web analytics measures what aspects of a website are working successfully towards a particular business objective. These business objectives could include a visitor purchasing a product, or a visitor signing up for a newsletter or even a visitor reading a blog.

Web analytics, which date back to the beginning of the web, have evolved from tools narrowly aimed at technology professionals into the broader domain of marketers, website designers and business executives. Gone are the days of simply trying to get visitors to a website. Today the focus has shifted to turning those visitors into customers. Armed with intuitive user interfaces and advanced data collection methods, today’s web analytics tools are helping website owners to measure the success of their online business goals.

Collecting the Data

The data provided by web analytics software is collected in two ways. The first is through the “log files” that your hosting company or in-house web server automatically tracks. These log files record numbers of visitors, which pages they viewed, when they visited your site, pages on which they exited and more. They also record the activity of search engine spiders, such as those from Google, Yahoo!, MSN and others.

The second method, which can provide much more detailed and accurate data, is called “page tagging”. Page tagging is required for analytics packages that are hosted on another server, and requires the insertion of additional code (usually JavaScript) on each page to be tracked. Such as a script sends the required analytics information to the analytics server, and often times will place a “cookie” on the user’s computer to provide more reliable information.

Both data collection methods are common, and, in fact, many analytics solutions make use of both. Either way, the analytics programs gather information about site visitors, distill it into easy-to-read reports and charts, which you can then use to make decisions about your ecommerce business.

Getting a Bird’s Eye View

Once visitor data has been collected, the next step is to look at the data and determine behavior patterns. Traditionally this has been a very intimidating proposal for most people, and a lack of understanding has led to difficulty in interpreting the information. In order to clear up some common misconceptions surrounding web analytics, the following are some terms that every online retailer should be familiar with:

• Hit: A hit is a request for a file from a server. A hit is registered for every file that is requested, whether it is an HTML file, an image, a CSS file, or a JavaScript file. Hits are in no way a measure of the number of visitors that view a site.
• Page View: A page view is a request for a file that is defined as a page in an analytics program. This can mean a request for an HTML document or a server side script, such as a PHP script.
• Visit: A visit is a series of requests from the same uniquely identified client within a given time period. This is usually set to 30 minutes, so that if a user waits an hour between requests, the server will register two visits.
• Unique Visitor: A unique visitor is essentially an individual coming to the website. Technically it is defined as a uniquely identified client generating requests on the server.
• Repeat Visitor: A visitor that has made at least one previous visit to a website.
• New Visitor: A visitor that has not made any previous visits to a website.

While these may seem like simple pieces of information, they can be used to extrapolate some very complex behaviors. For example, imagine an online retailer that is spending over one thousand dollars a month on a pay-per-click advertising campaign. The ads all lead to the home page of the retailer’s website, which shows that a lot of unique visitors are viewing that page in their browser. However, none of the other pages on the website show any traffic at all. Based on the simple information above from a web analytics solution, the retailer can assume that there is some sort of problem with their home page, which is hindering visitors from getting beyond that page. Perhaps the page has broken links, and the retailer was unaware of it. More likely is that the pay-per-click ads are targeted to a page that is not relevant to the ad, causing visitors to leave the site immediately. By simply pointing the PPC ads to a page that is much more relevant on the website, or changing the wording of the ad itself, the retailer can see more traffic throughout the website, as intrigued visitors begin to poke around more.

Visitors Aren’t Customers

In addition to displaying data about visitors, analytics solutions usually have some functionality for measuring business objectives. Any time that a desired, predefined action is fulfilled by a visitor, a “conversion” is said to have occurred. For most online retailers, seeing visitors convert to paying customers is the ultimate goal. However, there are other actions that retailers might want to track, such as a user signing up for a newsletter, or downloading a white paper. The aim of web analytics in this regard is to measure and report on the capabilities of a website to create conversions, which allows a retailer to better understand how visitors interact with the website.

Enter the “conversion funnel”, which is a common feature in many analytics solutions. A conversion funnel is a scenario of steps that a visitor must complete in order to fulfill a goal. It is called a funnel because as visitors move from the first step to the next, presumably many will drop off. The visitors that make it through to the last step represent “conversions” to that goal. Analytics software can help to identify “leaks” throughout the process by showing which steps the visitors are leaving.

For example, many ecommerce sites have a checkout process that spans multiple pages. Using web analytics software and a defined conversion funnel, a retailer can catch, say, that a lot of visitors make it through the first two pages of the checkout process, but leave the site after that. Upon scrutiny, the retailer discovers that the third page of the checkout process is not “secured”, and is also the page where the visitor is expected to enter their credit card. A quick fix to site (adding the page to the secure server) and visitors are making it through the checkout process just fine, and conversions are on the rise.

Another scenario is the case of a retailer using analytics software to review their marketing efforts. Offline advertising can direct users to individual landing pages, allowing the retailer to see what methods of offline advertising are working, and also to follow the visitors through the conversion process. The same process can be used to determine which online advertising efforts are driving the most traffic, and creating the most conversions. These online efforts can include email campaigns, PPC ads and banner ads. Finally, retailers can fine-tune their search engine optimization strategy by analyzing the content on their own website, and determining which keywords, phrases, and product descriptions, for example, deliver the most traffic from search engines.

Bells and Whistles

Bridging the gap between raw data and useful reporting has become a top priority for web analytics providers. As a result, there are many features and graphical interface innovations that are worth reviewing. One such feature, which represents a major step forward in the presentation of web analytics data, is called a “site overlay”. A site overlay provides a visual representation of a retailer’s website from the perspective of a web browser. The overlay can provide statistics, such as total clicks, unique visitors and conversion rates, for each link on the current web page. In addition, some providers are able to tie-in information about sales, such as total revenue per link and average revenue per click. The site overlay feature allows retailers to easy visualize how visitors are navigating their website, giving them an idea on visitor behavior and, as a result, how to optimize their site to increase conversions.

In sum, web analytics helps an ecommerce owner to improve his online business. The analytics programs review server logs and page tags to gather information about visitors to your site. That information is then condensed and summarized into reports, charts and graphs that an ecommerce operator can understand and comprehend. By understanding visitor behavior, website owners can have more than just a presence on the web. They can have a successful, growing business.