Google Analytics: What is a Filter?

Most people look at a bolt of fabric and see nothing more than cloth. A seamstress looks at it and sees a shirt. The data collected by Google Analytics is just about the same – it's only meaningless data until you view it from the right perspective. That’s when it begins to look like something useful.

In Google Analytics, a filter provides the right perspective. Filters help to separate data into two categories: the data that is used to create reports and that has no value to you. Google Analytics provides filtering capabilities that help you see through the myriad facts, numbers, and values it collates.

What is a Filter?

Suppose Google Analytics simply collected information about your website statistics and then dumped it in your lap without any kind of organization. It would take you longer to make sense of the statistics than it takes for a toddler to clean his room.

To help you understand what the facts that Google Analytics collects mean, data goes through filters. These filters can exclude information collected about certain domains or IP addresses (an Internet site’s numerical address), or they can simplify complex sets of numbers or facts, making them easier to understand.

Because understanding data can be a real chore, Google has created da set of MadLibs-like filters that give you the ability to separate your collected metrics by plugging in key pieces of information or patterns expressed in a language called Regular Expressions, also known as RegEx.

Some Example of RegEx:

• Exclude all clicks from a domain: this filter lets you exclude visits to your site from a specific Internet site. This is especially helpful if your company website gets a high number of visits from people on the corporate intranet looking for dirt to dish on their coworkers.
• Exclude all clicks form an IP address: Remember that girl who had a crush on you in third grade and now follows you everywhere, making untoward suggestions for weekend activities? She visits your website 150 times a day from her always-on cable modem connection 68.88.88.88. You could filter out that IP. But say she also uses her DSL connection, which had the IP address 68.88.88.89. this filter will also exclude information about visits that match a particular kind of pattern. Filtering on 68\.88\.88\.8[89] will match either 68.88.88.88 or 68.88.88.89 and will keep her obsession from screwing up your webs-site metrics. Now,if you could only use a filter to keep her out of your favorite restaurant.
• Include only traffic from a specific subdirectory: Your hot, new product finally has a page on your web site, and now you want to know how much traffic that one part of the site gets. Disappointing or not, this filter will show you how your baby is doing at the expense of all the other data on your site. “Include only” will include only the specific information that you tell it to.

In addition to these, you can create custom filters that separate out the information that you don’t want or that isolate the information you do want. custom filters allow you to:

• Exclude pattern: This filter will exclude data from visits that match a certain pattern. Say you want to collect information only on your catalogue’s regular products, not the sale ones. All sale products have a special E-commerce Item Code that begins with “SALE.” You could filter that field to exclude any hits from those e-commerce items by filtering with SALE.* as the pattern.
• Include a pattern: Just as you can exclude, you can choose to include information that matches a certain pattern. Say you want to measure only the visitors ho have really big screen resolution because you’re going to launch a new game that requires it. You could include traffic where the Visitor Screen Resolution matches d\d\d\dX\d\d\d\d, which would only match resolution with two four digit numbers.
• Search and replace: Much like the search-and-replace function in your word processor, this filter lets you search for specific types of information related to user visits and replace it with other information.
• Lookup table: Certain specific information isn’t in a format that you can understand, even if you have it in front of your. So the lookup table collects that information and translates it into a format you can understand. For example, you can use a lookup table filter to show readable names for unrecognizable or confusing URL patterns.
• Advanced: Do you wish you could exclude on pattern at the same time you’re including another? You can; just use an Advanced filter that can look at multiple pieces of information at one time.