Three Flavors of Web AnalysisWeb analytics provides three different levels of analysis. The first is a simple counter that tracks how many times something happens, such as how many people view a certain page or how many people enter your site through a specific link. The next level of analysis is ratios. Rather than merely counting, some simple mathematics are used to derive a deeper level of information from the counts. Ratios include how many pages are viewed per visit, what number of the total visitors exit after viewing only one page, and so on. The third and final level of analysis tracks the particular events that the user considers important. These are usually called key performance indicators (KPI) and they can be quite specific, such as the number of visitors to a page that enter an email to download a file, or broad, like the number of ad click-throughs per 1,000 visitors.
We’ll look at some important terms from each level, but because we're focusing on the basics of analytics in this article, we'll concentrate on the first two levels.
The Basics of Web AnalyticsThe main unit of measure for Web analytics is a page of content. Calling something a page can be a bit misleading, as we tend to think of a page of text. However, a page in this context just means the HTML page that holds the content, whether that content is text, video, a flash game or anything else you can host on a Web page. Web analytics collects data on the three main metrics on each page of content. These are:
- Page Views: A count of the number of times a page is viewed. With one page, this would be a count of the views on that page, but if there are multiple pages, the page count can be given by page or as a total of all the pages on the site.
- Unique Visitors: This is the number of people who come to the site and view the content - whether one page or many - during an analytical reporting period (month, week, day, etc.). Each visitor is only counted once during the reporting period, even if he or she returns many times.
- Visit Duration: Also called time on site, this metric measures the total time a visitor spends on the site - whether on a single page or many pages.
Ratios for Website AnalysisThere are many, many ratios used in Web analytics. Custom ratios can also be created. There are, however, some common ones that every site uses. These include:
- Pages per Visit: Also known as page depth, pages per visit counts how many pages the average visitor views. The more pages per visit, the greater the page depth - which is basically another way of saying the greater the user engagement with the site.
- Click-Through Rate (CTR): This ratio measures how many people clicked a link compared to how many people viewed it. CTR can be used to measure ad performance, A/B test related link setups, and so on.
- Bounce Rate: The bounce ratio is the number of single page visits divided by the total visits to the site. A high bounce ratio results in a low pages per visit and suggests that the surrounding content is not interesting or prominent enough to warrant a second click.
Moving DeeperAs mentioned, the KPI metrics that will be useful to a given website will depend on the site's objectives. The most general one is the conversion rate. This simply divides the number of users who carry out a desired action (email sign up, product purchase, ad click, etc.) by the total number of visitors. Again, Web analytics are flexible, so a KPI can be applied site-wide or to specific pages.
But there is much, much more to Web analytics than what we have covered here. Just to whet your appetite for more, there are analytics that:
- Identify the pages through which users enter and exit the site
- Rank external sites by how much traffic they refer to yours
- Show the flow of users through your pages
- Map which link placements get the most clicks
- Segment users into groups such as new, repeat and returning
- Provide demographic data on users