One of the most powerful ways to segment your data is to classify, or in other words, group, line items together. This gives you the power to compare different groups of data quickly and gain much batter insight into your trends and data. This section will cover what classification are and how you can add them and upload data to them.

  • Describe Classifications.
  • Understanding when to use Classifications.
  • Create Classifications in the Admin Console.
  • Use SAINT to upload and download Classification data.
One way to think about the various classifications is to think about a huge bag of candy. Within this bag we have hundreds of different pieces of candy. If we look into bag, we might be able to see a few types of candy on top, and choose one that we like. However , we probably end up mixing up the bag with our hand to try and see if there is something better. Now imagine a giant bag with hundreds of thousand of pieces of candy. You can quickly see that it would be hard to make good decisions with such a huge numbers of items to consider.

In this case, it would be very helpful to group together, or classify, our candies into appropriate groupings. Some of our candy classifications could be
  • Size (large, small).
  • Flavor (chocolate, sour, sweet).
  • Packaging (wrapped, no wrapper).
  • Ingredients (with nuts, without nuts).
  • Consistency (hard, soft).
Once we have created our classifications, or the ways that we want to organize and describe our candy, we can segment our candy into these groups so that we see how many candies are large versus small or chocolate versus sour. We can also then easily find the right kind of candy when we have a craving for it.

Keep in mind that one piece of candy could fit into multiple classifications. Foe example, a chocolate bar in its wrapper with nuts would fit into three of above classification.

When it comes to classifying you web data, the process is very similar (more so if you run an online candy store). You collect massive amounts of data, some of which are described very granularly, which is easily grouped or classified. This data could be your products, your campaigns or customer IDs, just to name a few. In any case, as you start to analyze your data, you will likely ask questions that require you to classify it. For example:

      “How did these kinds of products perform compared to those kinds of products?”
      “Which group of campaign links drove the most success on our site?”

Once you start looking at your data this way, you’ll want to classify as much as possible in order to gain additional insight info into your Web analytics data.

Example 1: Products
Let’s first assume that we are an Electronic Retail site on the Web. We have thousands of products, and when someone purchases a product on our site, we have the code in place to capture that product and send it into SIte Catalyst. By running a Products reports report in the Commerce area of Site Catalyst, we can see which products sold (or are selling) the best for our chosen time period.

The only problem with this is that we don’t have just a few products to look at. We have thousand of products, and although we can look at the top 50 or so and see which specific products are selling the best, it is possible that there is a group of products selling well as a whole that we we are not seeing because because they are not very high on the individual list.

Therefore, it becomes very valuable to put products into groups. In fact, while analyzing this product data, we might already be asking questions similar to this:

      1. How are different manufacturers performing?
      2. How are different product groups performing?

To answer the first question, we want to know if our Apple products are outperforming our HP or Dell products. For the second questions we want to know how PDA sales are compare to Desktops and Monitors.

Technically (and if we wanted to go crazy) we could look through every line item in the report and add up the numbers every time we asked these questions. However, classification allow you to group these line items based on the questions you are asking and quickly get the numbers you want when you run the report.

For example, after going through the classification process, which we will discuss next, you could have a Manufacturer report that would add up the revenue for all of the products in real time. In addition to the Manufacturer report, you could create any other Classification reports that would help to understand the products on your site, Like Department, Sub-department, Style, Size, etc.