Google Analytics Profiles and Filters Part 2: Advanced Filters

This is Part 2 of a two-part series on Google Analytics Profiles and Filters. Part 1: The Must-Haves, covered what I consider to be the basics. Now we’ll get into some of the more advanced, and interesting, GA filter techniques.

Note: Since I posted Part 1, Google Analytics has released their beta versions of Advanced Segments and Custom Reports. These tools are pretty incredible, and can potentially be used in place of many filters. The decision to use profiles and filters vs. advanced segments will depend on each company’s logistics. It will be a case-by-case basis and is something I will cover in a future post.

Advanced Filters Defined

Advanced filters let you extract data from any of the fields in Google Analytics and manipulate that data.

In the example above, we are extracting the Campaign Source from Field A, and the Campaign Content from Field B, and then we combine them together into the User Defined field. Because the radio button for Override Output Field is set to Yes, our GA reports will now populate the User Defined field with our custom data.

You’ll notice something funky in the Extract A and B text boxes: (.*). Advanced filters let you use regular expressions to extract data from the field. I’m not going to get into regular expressions. They are hugely powerful, but very complicated. (.*) basically means grab all of the data from the field.

The Output To field combines the data from Field A ($A1) and the data from Field B ($B1) and separates them with a hyphen.

Even More Power - Cascading

I think you can see the power of even the above simple implementation of advanced filters. Using this technique you could easily correlate a Campaign Term with an E-Commerce Item Name for example. Then you could create a report showing what your visitors purchased when they searched on a specific term. The options are endless. It’s up to you to decide what you want to measure.

Where these filters get even more powerful is when they are used in a cascading fashion.

You can actually extract data from two fields and combine that data into a custom field. Then you can filter on that custom field to further customize your data.

This can get confusing, so I’ll give a specific example of how this might be used in the real world to create meaningful data.

NY Times Online Marketing Campaign

Your company is running a huge online marketing campaign on the NY Times web site. You are running tons of ad variations across multiple pages and want to generate reports on which specific ads are driving the most, and the least, traffic.

Every time a visitor clicks on one of your ads, you are able to populate the various Campaign fields within Google Analytics by using Link Tagging.

There are rules and specific variables to follow and use when generating your link tagged URLs.

Here is an example of what the code might look like for your “link tagged” ad:

http://yoursite.com/page.html?utm_source=NYTimes&utm_medium=Banner&utm_content=BlueAdWhiteText&utm_campaign=Fall09

If you look at this closely you’ll see we are capturing that the click came from the NYTimes site (source), it was from a Banner (medium), the ad variation was BlueAdWhiteText (content = ad content) and it was from the Fall09 campaign (campaign).

All of this information gets pulled into GA because the destination URL is link tagged properly.

Filtering the Data

Now that all of your ads have been link tagged, we can set up our advanced filters to combine meaningful data on which we can then run reports.

Let’s put the Campaign Source, Name, and Content together and place it in the User Defined field. Then we’ll be able to pull up the User Defined field in any report and see how many visitors are driven to your site by each specific ad (Content) from the NYTimes Fall09 marketing campaign.

Filter 1

This first filter extracts the Campaign Source in Field A and the Campaign Name in Field B and outputs them together into Custom Field 1. In this case all data that comes from your link tagged URLs would be output to Custom Field 1 as: NYTimes - Fall09

Filter 2

The next filter takes Custom Field 1 from Field A, combines it with the Campaign Content from Field B and outputs the results into the User Defined field.

Now you can easily see, in one location, how many visits you received from each specific ad running in your NYTimes Fall09 marketing campaign by running reports on the User Defined field in GA.

Endless Possibilities

As I stated earlier, there are endless options when it comes to applying filters to all of the different data fields. You need to decide what will be the most useful for your company and go for it.

With that said, be VERY CAREFUL.

Once you filter data, everything you filter out is gone forever. Because of this you should ALWAYS create separate profiles on which to apply filters. Always keep one profile with no filters. This profile will have all of your raw data, just in case you make a mistake with the filters. (See Google Analytics Profiles and Filters Part 1)

Dive in and start testing.

View related topics: google analytics, google filters, google profiles