If you work at an agency, and you are fortunate enough to have clients like mine that internet marketing savvy, I hope this post will help start some interesting conversations, and help open the door to additional tracking that you might not yet be utilizing with Google Analytics.

The other day, one of my more hands-on clients forwarded me this article about monitoring rankings using advanced filters ^[1]. After reading it I knew that I would be able to use this for all of my clients, and I started telling my colleagues about how helpful it would be for everyone to use. However, I forgot one important detail. Not all of us know about filters, let alone advanced filters. I also realized that we haven’t really implemented very many “User Defined” segments, so all in all, we’d have to start with those steps first before we truly appreciated the “golden nugget” that had been uncovered.

What are Google Analytics filters & how are they used?

There are two types of filters, predefined and advanced. Google describes them as:

• Predefined Filters: Predefined filters are a quick and easy way to accomplish some of the most common filtering tasks.
• Custom Filters: Custom filters allow more advanced manipulation of data.

So what does this mean for you, and how are they set up?

Note: When working within an account it is a good idea to always have one profile without filters. It’s important to keep a “clean” profile, and create additional profiles that contain segmented data. This will help ensure the integrity of your data in case a profile is deleted or a filter is set up incorrectly. Just imagine if you only tested on your main profile – you’d have no way of knowing what data was accurate, and if you deleted the profile, you’d have no way of getting historical data back.

Creating Predefined Filters

There are a few choices when creating a predefined filter, so you first have to decide what you want to exclude or include in your profile data. After you know what you want to track, log into your account, and choose the profile you will be working with (or create a new one). Once you are in the profile you can add/edit filters from there.

1. Exclude all traffic from a domain. Use this filter to block all traffic from a certain domain. Google help explains that Google Analytics uses reverse IP lookup to find the domain of your users, so this can come in handy for larger companies that have their IP addresses mapped to the domain instead of their ISPs.
2. Exclude all traffic from an IP address. We use this one all the time to block traffic from our IP address. This helps ensure that the testing I do for clients, and the visits I make to their site doesn’t skew the data. I know the level of traffic I generate isn’t all that much, but by excluding our IP the data is just that much cleaner, and who doesn’t like that?
3. Include only traffic to a subdirectory. This helps you track only the traffic that goes to a subdirectory of your site. This is particularly handy when you have a blog that you want to monitor traffic on. You would simply enter ^/blog/ if you want to track www.mysite.com/blog.

Creating Custom Filters

There are a variety of custom filters that you can create, each one giving you a deeper dive into the data that is captured in your Analytics account. When creating a custom filter, you first have to choose one of the following:

• Exclude
• Include
• Lowercase
• Uppercase
• Search and Replace
• Advanced

From there you will make a selection in the filter field that will help further segment your filter. There are a variety of choices ranging from Request URI, data from your campaigns (like keywords), eCommerce data, and visitor information. When paired with the choices above, you can see how quickly you will be able to create an immense number of filters to help you really hone in on certain parts of data.

User Defined Segments

If you checked out the article I referenced above, you’ll know that the other feature mentioned and used in the advanced filters they created were user defined segments. You may have seen these when viewing reports in Analytics, either on the top under advanced segments, or on the left hand side under Visitors. Either way, it’s important to understand that they are just one more powerful tool that we can use to get a better idea about who is visiting a website, and how one segment’s behavior may be different than another’s.

Still wondering why you should care? User-defined segments allow you to label your visitors based on the actions they take while on your site, such as sign up for your newsletter, fill out a contact form, or make a purchase. And considering the fact that all of these actions would lead us to believe they are probably more engaged than the average Joe, it would be really great if only we could see how they interact differently when it comes to other important metrics like time on site or pageviews, or even bounce rate.

Setting up User Defined Labels

Now don’t leave when I say this, but user-defined labels require a bit of extra code. But it’s not that bad, I promise! Firs you have to decide the name for your segment, such as “customer”, and you add a little snippet of code that helps identify that label when they reach the confirmation page. Check out this help section for specific details on adding the user-defined code ^[2]. (Be sure to scroll down to the “technical notes” to make sure you understand all aspects before diving in.)

So now that you understand filters and user-defined labels, you can start to do some of the experimenting that Patrick discusses in the article above, and start gathering some pretty amazing data that will help you better optimize your accounts.

A friendly tip:When dealing with keyword reports or filters, you are bound to notice that there are visitors coming in through (not set), and it might be a fairly large number. This is because not all visitors come in from a keyword; some come in directly either by typing in your URL, or through a bookmark. You can verify that your filter is correct by calculating the percent of visitors that came in through (not set), and compare that to the percent of Direct visitors in the Traffic Sources report. They won’t be identical, but they should be fairly close.