With the insertion of iPad, recently announcing over 1 million sold, and the growing base of smart phone users I wanted to highlight the quick resources you have available in Google Analytics to view this level of data.

Under advanced segments there is a default segment for Mobile Traffic. This includes traffic from iPad, iPhone, iPod, Android, BlackBerry, Windows, Samsung etc.

Under the visitors section on the left hand side navigation is a drop down for Mobile which includes two sections 1) Mobile Devices and 2) Mobile Carriers

Explore and learn from the data, if you’re seeing traffic and conversions you may want to investigate tapping into this market.

It is imperative to utilize Web Analytics to make informed, actionable decisions on adjustments to your site to increase conversions. A great tool recently added to Google Analytics is the option to add annotations. This enables you to make notes of changes you have made to the site within the Google Analytics interface, so you can easily see what and when those changes were implemented, so when analyzing the impact of them you can quickly pinpoint which change(s) were effective. For those of you not using this feature I would suggest you jump on the wagon, it’s extremely useful.

Samantha Bedford, Location3 Media’s recently-dubed SVP in charge of account services, discusses web analytics. Samantha talks about why analytics is so important for online marketing campaigns, which analytics provider she prefers, the new features released by Google Analytics, interesting projects she’s worked on and much more.

At the top of my GA wish list has always been annotations – a place to document changes and make notes. Well, Google answered that wish in early December by including annotations, allowing you to place a quick note when you click on an actual date. This feature was not automatically available to the public; however, earlier this week we spotted it in our accounts. Some accounts had it and some didn’t, so we did some research to figure out why. Here is our conclusion:

Under 10,000 visits: Annotations

Above 10,000 visits: No Annotations

The accounts that had less than 10,000 visits this month had annotation capabilities while the others did not. The only reasoning we can come up with for this discrepancy is scalability. If Google tried rolling this out to every account at once and traffic suddenly dropped it would be easily noticeable and they would receive a giant web backlash. It’s easier to test new features on a small scale; if small drops appear, it’s less noticeable. Now i’m just waiting for Google to roll it out to my other accounts and i’ll be able to document all the changes that happen to a website.

Update: 1/27/10, Annotations have spread to all our of accounts.

Like the majority of people everywhere, I hate filling out long forms riddled with seemingly useless fields. Like a small minority of people, I hope and dream of the day when website creators will come to their senses and create smaller forms. To help make this day come true, I did some research and learned, through some extra Google Analytics work, webmasters can install code and to determine where people left the form. All that needs to be added to the form field is a virtual pageview. By determining where people leave the form, you can eliminate fields that people are loathe to complete, thus shortening the form and increasing conversions.

I’ll try to demonstrate what is needed to create a form field virtual pageview for the current and asynchronous google Google Analytics. One of the most important things you can do while setting this up is create a goal and funnel. Each specific field will contain its own entry that is entered in the funnel. (see image below). The funnel allows you to see where people left and which page they went to.

(There’s more after the image)

Coding

Form

<form action="cgi-bin/formhandler.pl" method="post"></form>
 
Current Google Analytics Code
<pre lang="javascript"><input name="formfield" type="text" />

New Asynchronous Google Analytics Code

<input name="firstname" type="text" />

The coding needs to have various elements for this to work properly, the most important of which is the form name. The form name starts the process of the javascript event to track everything properly. After the form name, you need the name of the form field which is found next to the field “name.” After this is established, a normal virtual pageview is placed.

Even though the codes look different they have the same basic functions with the javascript onblur event. The onblur event is triggered when someone begins typing in the field—the act process of being in the field is sent to Google Analytics. Don’t worry, the information isn’t captured; it just records that someone typed in the field. After the form fields have been completed, a normal virtual pageview is used for the submit button. This can be found in my previous blog, Creating Virtual Pageview with Google Analytics Asynchronous Tracking.
If you have any questions or concerns feel free to leave a comment. ~Lorrin]]> http://www.expertsem.com/2010/01/21/how-to-create-an-asynchronous-virtual-pageview-for-form-fields/feed/ 0 http://www.expertsem.com/2010/01/19/am-i-drinking-the-google-analytics-kool-aid/ http://www.expertsem.com/2010/01/19/am-i-drinking-the-google-analytics-kool-aid/#comments Tue, 19 Jan 2010 22:19:23 +0000 Alex http://www.expertsem.com/?p=2795

Drinking the Google Kool Aid

Location3 Media has been promoting and utilizing Google Analytics for a few years now and the solution has been a great source of information for a majority of our client base. We certainly haven’t missed the expensive paid tools that were typically implemented inproperly and didn’t provide any real value.

As we take the time to look over every piece of our marketing programs I am personally beginning to fall off the Google Analytics bandwagon in favor of (GULP) Yahoo’s product, formerly known as Index tools.

Take a read over this great comparison between GA and YWA, perhaps slightly biased but good info none the less.

The next step internally is to do the same type of side by side testing between the two offerings and develop some of our own conclusions. So I guess I’m not drinking the Kool Aid any longer…. Any insight from any readers out there?

As I was updating my smart phone advanced segment the other day I was surprised to find that my client was also getting traffic from two new breeds of internet browsing tools. The first, the game console. Of course! Playstations and Wiis can be a substitute for computers. The second, the portable music player such as the Sony X1000 Walkman, has internet browsing capabilities. I immediately rushed into my advanced segment edit area and started creating. I needed to know how this traffic was performing! The results ladies and gentlemen, were neat.

Client #1:

Client #2 (I went a step further in this one and broke out paid as well):

Google Analytics-Specialty Segmentation

Although the traffic levels weren’t earth shattering, I was surprised to see in both clients that the traffic sent from these segments had high engagement rates. I am assuming we will see the traffic from these segments continue to grow as the year goes on. In the meantime, remember, that it isn’t just computers and phones driving users to your site anymore.

So everyone, including myself, was excited about the new type of engagement goals and the increased number of goals in Google Analytics. No question, they are very helpful. With this new feature you can now quantify the portion of your visits based on their level of the engagement with your site. However, if you have used this new feature, you are probably starting to notice some strange issues.

Here are the new features:

• 4 sets of goals with 5 goals in each set, thus up to 20 goals available
• 2 additional goal types: time on site and pages/visit goal; also called threshold goal or engagement goal

The problem

You would think it is very logical, but wouldn’t it be nice if you could set up multiple levels of engagement goals since Google is offering 20 available goals out there?

If you want to find out how many visitors spent at least two minutes, at least four minutes, at least ten minutes and so on, and if you want to find out the same about the number of pages viewed within each session, you must set up goals for at least two pages, at least four pages and at least ten pages for each session.

I agree it is straight forward if you are ready to check out how engaged your traffic is at different levels.  But think about this:  What if one visitor spends ten minutes and views ten pages during one session. Then you would find one visit with a 600% total conversion rate. This is because Google counts the goal three times for each type of the goals. In theory, Google is right in terms of the goal number; however,  the total goal conversion rate is very confusing.

The other problem with the new goal set is very obvious¾Google has not updated its custom report for goal 5 to goal 20. Its default interface will allow you access to those goals; however, if you try to set up your own custom report with more than four goals, you are going to need some manual calculation for the additional goals.

The best practice

• If you can, only use one engagement goal for each type
• If you like to measure the different levels of engagement, don’t use the total goal conversion rate metrics, and only look at the performance of its individual goal

A standard practice in any analytics package – not just Google Analytics, which is our preferred solution – is to block traffic coming through from your own IP address(es). This ensures that the data collected and utilized to develop actionable insights is pure and not skewed by data coming through from people designing your website, an agency or people testing the site.

I wanted to not only remind you to put this vital filter in place when you set up analytics, but also remind you that there are two types of IPs – internal and public.

In theory, the only way the internal IP would show traffic on your analytics reports is if the IP is on the same LAN as the site’s server. Every external visitor to the site should only be reporting their public IP address.

Naturally keep this in mind if you are running analytics on your own sites, which reside on the same LAN as the site server. If this is the case you will want to set up two profiles in Google Analytics, the first for all traffic to the site blocking public IP addresses, the second profile blocking public and internal IPs.

Any questions feel free to add to this string.

Landing URL tagging refers to adding special tags at the end of a landing page URL.  Those tags contain variables, such as campaign name, medium, term, content and so on. Without tags, an analytics package treats all traffic—organic and paid the same.

If you are only concerned with cost data or click data (i.e., impression, cost, CTR, etc.), then you do not have to worry about your tagging strategy. You can find that information on your CPC platform (e.g., AdWords). But if you want to find out visitors’ behavior after they click on your ads and visit your site, behavior such page views, bounce rate and time on site, you must use an analytics package like Google Analytics and tag your campaign correctly.

Google Analytics offers both auto and manual tagging options. With auto tagging, you don’t have to do anything other than make sure the auto tagging setting is on. Google automatically adds GCLID tags without any manual process. This code is unique at a keyword level.  The resultant URL looks like this: www.mysite.com/?gclid=123xyz. But auto tagging is only for Google’s AdWords campaigns. Therefore, if your clients are only running campaigns on Google, we would recommend this. It is easy to set up and it is more accurate.

If your ads are across multiple engines, you would have to manually tag CPC campaigns for other engines. Manual tagging requires appending campaign variable tags to the end of URLs. You can use Google’s URL Builder tool to do this. Here is an example of the manually tagged URL:

If you do manual tagging, here are the two common problems faced:

1. Google Analytics does not give ad group level information. It only shows the campaign- and keyword-level data.
2. In order to identify the paid keywords, you must fill in each keyword in the “Term” field in the URL Builder tool. If you have a big account with millions of keywords, this could be a very complicated process.

We recommended the following solutions for these two problems:

1.Use a unique ad group-level identifier for campaign names. For example, you can use campaign name and ad group name together to represent the ad group.

Campaign name: all product

Ad group Name: discount product

Campaign name to tag: all_product_discount_product

In this case, by looking at your campaign name, you would be able to see ad group-level information, within your campaign level reporting.

2. Use a dynamic key to catch the actual keywords

A dynamic key is a word that picks up the actual keywords. By using a dynamic key, you are able to save the actual step of putting in the real keywords. We have tested some dynamic keys for ads on major search engines on our own campaigns. The result shows the following three work the best.

• Yahoo: use {YSMKEY}
• Bing: use {Keyword}
• Google: use {keyword}

By using a dynamic key, your URL tag will look like this:

That’s it for this quick tag strategy. Please keep us posted if you have found other solutions or have any questions regarding the above content.