Survival in hard economic times means concentrating on what works and ridding yourself of what does not. Without a firm grasp on how everything is working, you would not be able to shift resources to assets that are paying off and away from those that are failing. That goes for everything, including your company website.

The tools and information you would use to see how your website is doing are collectively known as Web Analytics. They focus on the measurement, collection, analysis and reporting of data for the purpose of understanding and optimizing website usage and they come in two varieties: On-site and Off-site.

On-site web analytics measures the performance of your website once the visitor reaches your site. This data is typically compared to key performance indicators, a set of values used to measure results against, and used to improve the website’s audience response. Off-site web analytics, on the other hand, measure potential audience, visibility and buzz around the Web. This kind of measurement can be performed on nearly any site, whether you own it or not.

The Language of Web Analytics

If you are going to be analyzing the performance of your website, you need to know the language of that analysis. According to the Web Analytics Association (WAA), there are three basic types of metrics—counts, ratios, and KPIs:

Count. The most basic unit of measure; a count is a single number, not a ratio.

Ratio. Typically, a count divided by a count, ratios can use either a count or another ratio as the numerator or denominator.

KPI. Key Performance Indicators are frequently ratios, but they can be counts as well.

There is a fourth type of definition that describes concepts instead of numbers, Dimension, which can be interpreted like counts, but are usually further qualified or segmented to be of any real use. Metrics are measured across dimensions. Because of this, dimensions define a more general class of metrics and represent data more closely associated with individual visitors.

The WAA has also identified what they call “universes.” These are aggregate, segmented or individual and a metric can apply to any one of them.

Aggregate. Total site traffic for a defined period of time.

Segmented. A subset of the site traffic for a defined period of time, filtered in some way to gain greater analytical insight.

Individual. Activity of a single Web visitor for a defined period of time.

Now, with the basic measurements down, it is time to move on to the various data being measured:

Hit. A request for a file from the web server.

Page View. A request for a file defined as a page in log analysis or the script being run in page tagging. One page view may generate multiple hits as all the resources required to view the page are also requested.

Visit / Session. A series of requests from the same uniquely identified client with a set timeout.

First Visit / First Session. A visit from a visitor who has not made any previous visits.

Unique Visitor/User. The uniquely identified client generating requests on the web server or viewing pages within a defined time period.

Repeat Visitor. A visitor that has made at least one previous visit.

New Visitor. A visitor that has not made any previous visits.

Impression. An impression is each time an advertisement loads on a user’s screen.

Singletons. The number of visits where only a single page is viewed. The number of singletons can indicate “Click Fraud,” calculate bounce rate and sometimes identify automatons (“bots”).

Bounce Rate. The percentage of visits where the visitor enters and exits at the same page without visiting any other pages on the site.

% Exit. The percentage of users who exit from a page.

Visibility Time. The time a single page is viewed.

Session Duration. Average amount of time that visitors spend on the site each time they visit.

Page View Duration. Average amount of time that visitors spend on each page of the site.

Depth / Page Views per Session. Depth is the average number of page views a visitor consumes before ending their session.

Frequency / Session per Unique. Frequency measures how often visitors come to a website.

This is by no means a fully comprehensive listing, but it will get you going. For a full list and description of these data, visit the Web Analytics Association at their website www.webanalyticsassociation.org.

Analytical Technology

Now, with some idea of what to look at, it is time to examine the systems that you would use to work with this information. The choices are many and varied and going through them all to find the best fit for you can make your head spin. However, if you take it easy and do your research, you should be able to manage it.

Writing for CIO Today, Jennifer LeClaire, points out that web analytics software essentially falls into three categories: Server-side, Client-side and Hosted, and that this is where you should begin. She also says that you ought to consider your budget, what your Web hosting company supports, and your appetite for statistics.

Server-Side Web Analytics

Server-side analytics tools are applications installed on your website’s host server by your hosting company and accessed through your website’s control panel. They record each time a user pings your server, tabulate data on when your site was visited, where the visitors came from and the pages they viewed. These tools offer anywhere access, high reliability and relatively low cost, with most basic server-side tools costing nothing, while more powerful tools commanding higher prices.

Data security and privacy are other advantages of server-side web analytics tools. This is because the data is stored in-house. Also, since these tools record pings, they offer better information on broken links and other site failures. The downside is portability. If you change Web hosting services you could lose your information.

Client-Side Web Analytics

These are desktop applications, installed on your computer, that allow you to call up the data and access statistics with or without an Internet connection. You can also keep your original log files and store them yourself, which offers you full control over your data.

Client-side tools can be expensive and their cost is paid up front. A basic tool may run $500, but more robust enterprise versions would cost more. Still, for those who don’t want to be stuck with a given host, this can be an attractive option. Client-side software also allows you to manage multiple domains and it integrates well with existing enterprise software.

Hosted Web Analytics

Hosted Web Analytics are also referred to as on-demand Web analytics solutions. These are software-as-a-service (SaaS) applications installed on a vendor’s server. To use these tools, you log into to the Application Service Provider’s (ASP) website to view and analyze your data. That is about it. There’s no downloading, installation, or maintenance, which leaves you free to analyze your data.

Hosted applications are not free, nor do they generally come with high, up-front costs. Hosted services usually charge a monthly fee; and while that means a recurrent line-item on your budget, it will likely be something you can afford. Another advantage is that you will be able to access your data from any computer with an Internet connection, giving you a great deal of flexibility. Also, you can capture only the data for the pages you want to track. Of course, you have to stay on top of the pages because without server-side analytics to back up your hosted services, you will miss vital data.

Putting it All Together

At the end of the day, you want the best fit of capabilities and often that means using multiple technologies together to achieve the results you desire. For example, by building your base with server-side tools, you will have a good way to draw data that is based on tracking visitor pings to your server. These can include error messages, spider activity and other, similar metrics. Then, by adding client-side or hosted solutions to your server-side tools, you can look at your data in different ways based on the nature and capabilities of the client-side or hosted solutions you decide to use. Technology, though, is only half of the issue.

Remember, while all of the data is useful to a degree, they are not all equally useful. You should concentrate on the metrics that deal with what is important to you, events that equate to actions that you want to see your site visitors take, such as watching a video or opening your online catalog. These are the key things you want to know about, important events taking place on your website, so make sure your analytics makes these data clear and easy to find. Your decisions will be easier to make and their outcome will be better.