The true custom reporting flexibility available in WebTrends Analytics allows our customers the ability to report upon a number of different initiatives via the unique combinations of dimensions, measures, and filters. By providing a robust set of options at not just the dimension and filter level, but the measure level as well, it is possible to truly customize what is tracked and reported to your user base. My intention over my first few blog posts will be to provide a number of simple techniques that can be utilized and combined for such initiatives at your firm.

To use a real world example, I am going to focus upon tracking marketing efforts on social media/networking sites such as YouTube, Facebook, Twitter, Flickr, MySpace, and del.icio.us. As these and other similar sites have become a popular and cost effective way to market to both potential and existing customers, the ability to report upon the results of those efforts has become increasingly important.

In this post, I will cover the creation of a relatively simple report that uses a single dimension of Visit Duration. Prior to designing the report, we will construct a number of custom measures that will group our various Web 2.0 Campaigns into different buckets that we can then report on (in this case we will just look at Twitter and Facebook). These individual Campaigns could of course always be broken out to focus upon attribution, but the real goal of this basic example is to instead highlight some of the available flexibility when it comes to customized measures, the use of which can be quite beneficial when it comes to segmentation and high-level reporting for various user groups.

Step 1: Define Your Tagging Strategy

The first step will be to define your tagging strategy. In this case we are going to utilize the standard Campaign query parameter (WT.mc_id), sending a three letter code followed by an underscore and then a numerical code. This code should be unique per Campaign and, if later used in dimensional reports based upon WT.mc_id, could always be translated into a more user friendly string by using a standard look-up table. In this case, we will use the following naming conventions:

1.  All posts from Twitter will start with TWT_ followed by the Campaign’s unique numerical code.

2.  All posts from Facebook will start with FBK_ followed by the Campaign’s unique numerical code.

Step 2: Create Your Custom Measures

Now that our tagging strategy has been defined, we can begin to focus upon the actual implementation work and build our custom measures. I will call the first measure “Twitter Visits” and it can be created by going to the Measures section within the WebTrends UI, selecting New, and then choosing the following options:

General:

o Name: Twitter Visits

o Category: Marketing

o Column Name: Twitter Visits

What to Measure:

o Value to Base On: Query Parameter

o Parameter Name: WT.mc_id

o Parameter Can Have Multiple Values Delimited by Semicolons: Unchecked

When to Measure:

o When: Hits That Match Specified URL

o Do You Want To Sum This Measure Across The Visit: Yes

o URL Expression: *

o URL Parameter (Click New):

• Parameter Name: WT.mc_id
• Match Records That Do Not Contain This Parameter Name: Not Checked
• If The Parameter Is Present, Match Records Where: Checked
• Parameter Value: Equal To TWT_*
• Text/Numeric/Regular Expression: Text

Format:

o Currency Symbol: No Currency

o Decimal Places: 0

The next custom measure will be called “Twitter Page Views” and it will be used to report on all of the Page Views generated via our Twitter Campaigns. For this measure you will use the following options:

General:

o Name: Twitter Page Views

o Category: Marketing

o Column Name: Twitter Page Views

What to Measure:

o Value to Base On: Query Parameter

o Parameter Name: WT.mc_id

o Parameter Can Have Multiple Values Delimited by Semicolons: Unchecked

When to Measure:

o When: Hits That Match Specified URL

o Do You Want To Sum This Measure Across The Visit: No

o URL Expression: *

o URL Parameter (Click New):

• Parameter Name: WT.mc_id
• Match Records That Do Not Contain This Parameter Name: Not Checked
• If The Parameter Is Present, Match Records Where: Checked
• Parameter Value: Equal To TWT_*
• Text/Numeric/Regular Expression: Text

Format:

o Currency Symbol: No Currency

o Decimal Places: 0

After configuring the two Twitter measures, we simply create two more similar Facebook measures that use FBK_* as oppose to TWT_* within the URL Parameter section of the setup.

There are a few important things to note in regards to the above measures:

1.  How we choose to calculate these measures (count, average, etc.) is determined once the report is built. Although we have created two Page Views measures for Facebook and Twitter that we intend to base upon the actual counts, you could easily choose to report on something like Average Page Views within your custom report using the same measures.

2.  The only real configuration difference between the Visit and Page Views measures, other than the names, is the “Do You Want To Sum This Measure Across The Visit” setting. This, when coupled with the count option within the custom report itself, is what allows us to differentiate between Visits and Page Views. If you want more specifics regarding the setting there’s a great post on WebTrends Outsider.

Step 3: Create Your Custom Report

Now that we have all four measures in place, we will go ahead and create the report itself. This can be done by going to the Reports section within the WebTrends UI, selecting New, and then choosing the following options:

General:

o Name: Social Networking Traffic by Visit Duration(Twitter & Facebook)

o Title: Social Networking Traffic by Visit Duration(Twitter & Facebook)

o Category: Marketing

Dimensions:

o Primary Dimension: Visit Duration

Exclude activity without dimension data: Unchecked
Correlate multiple values with multiple values in measures: Unchecked (N/A)

o Secondary Dimension: None

Measures:

o 1st Measure:
Measure: Visits
Column Name: Default
Allow Column To Be Sortable: Checked
Disable Column Total In Report: Unchecked
Include Interval Data: Checked

o 2nd Measure:
Measure: Page Views
Column Name: Default
Allow Column To Be Sortable: Checked
Disable Column Total In Report: Unchecked
Include Interval Data: Checked
Method: Default (Sum/Can Also Use Count)

o 3rd Measure:
Measure: Twitter Visits
Column Name: Default
Allow Column To Be Sortable: Checked
Disable Column Total In Report: Unchecked
Include Interval Data: Checked
Method: Count

o 4th Measure:
Measure: Twitter Page Views
Column Name: Default
Allow Column To Be Sortable: Checked
Disable Column Total In Report: Unchecked
Include Interval Data: Checked
Method: Count

o 5th Measure:
Measure: Facebook Visits
Column Name: Default
Allow Column To Be Sortable: Checked
Disable Column Total In Report: Unchecked
Include Interval Data: Checked
Method: Count

o 6th Measure:
Measure: Facebook Page Views
Column Name: Default
Allow Column To Be Sortable: Checked
Disable Column Total In Report: Unchecked
Include Interval Data: Checked
Method: Count

Note that the standard Visits and Page Views measures were also added. This gives us insight into the site/profile traffic as a whole, while also giving us the ability to later add meaningful calculated measures.

Before moving on, here is an important reminder when it comes to the setup of similar reports that are based upon different dimensions:

Use caution when utilizing custom Visit measures, as we are summing across a Visit. If you use a dimension that is hit based (for example Content Group), single Visits could be counted multiple times. While this isn’t necessarily a bad thing, when choosing such a combination you will usually want to check the “Disable Column Total In Report” option. As my basic example uses Visit Duration, this is not necessary.

Step 4: Apply The Report To Your Profile(s)

The report can now simply be added to any of your desired profiles for testing purposes. Note that there should always be some QA involved when creating any custom filter, measure, dimension, or report. As this is a rudimentary example, it is not neccesarily representative of the number or order of steps that would be used in your case. Most firms will usually run through numerous testing cycles prior to report promotion or the production push of customized tagging.

Step 5: Implement Your Tagging Strategy

The next step will be to implement the tagging strategy discussed earlier and to begin sending the relevant data via your social media/networking postings. While this can be done in a multitude of ways, two of the easiest are as follows:

1.  Use a vanity URL such as facebookvi.webtrends.com, with each vanity URL redirecting the user to the correct landing page, while also adding the relevant Campaign ID. As an example, the above vanity URL could forward the user to:

http://www.webtrends.com/Products/WebTrendsVisitorIntelligence.aspx?WT.mc_id=FBK_00000001

2.  Use an online URL shortening tool such as www.snipurl.com, www.budurl.com, www.tinyurl.com, or www.urlborg.com to convert your landing page link (with attached the Campaign ID). Using urlborg, this example:

http://www.webtrends.com/Products/WebTrendsMarketingWarehouse.aspx?WT.mc_id=TWT_00000005

Becomes this:  http://ub0.cc/3x/1P

Step 6: Actionable Analysis

While the simple techniques outlined above can be utilized in a number of different and creative ways, as with any measurement and reporting strategy, the ultimate goal should always be to ensure that you are able to use the results from your analyses to help drive continous improvement.

While WebTrends Analytics provides an extremely flexible custom reporting interface, giving our customers the ability to gain true insight via the design and creation of tremendously diverse reports, always make sure to dedicate enough time to define the metrics that allow you to meet your site and organizational objectives.

The real sign of true analysis is the development of recommendations that can be acted on.

If you are using similar methods or have some thoughts on any of the topics covered in this post, we would love to hear from you.

Tags: actionable, custom reports, facebook, marketing campaigns, social media, social networking, twitter

This entry was posted on Tuesday, December 23rd, 2008 at 7:00 am and is filed under Best Practices, Emerging Technologies, Inside Webtrends, Marketing Optimization. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.