SEO Lion’s Blog » Web Analytics

Web Analytics JS code placement – Top or Bottom of a page?

seolion — Mon, 31 Aug 2009 15:48:06 +0000

This is one area many get confused – where should the web analytic JavaScript code to paste should go on a page.

There are many places where it can be placed actually, and in fact they are correct to the scenario for that particular page!

Let me explain each placement location in detail

Bottom of the page, just above closing tag: This is the most widely used placement for JS based web analytic solutions. It has its own advantage and disadvantage.Your page might contain a number of elements. The sequence on which these elements are loaded are based on the position at which they are included on a page. The elements (images, scripts etc) on the top portion of a page load prior to the one at the bottom.Consider that you have placed the web analytics code on top portion of the page immediately after opening tag and assume that you are using Google Analytics. Since Google Analytics is free, many users will be using it, and thus provides a heavy load on the server most of the time. Google is a great company who know to tackle these situation, but consider a worst case – server is overloaded and the requests are queued at the server. Your web page will wait for the Google Analytics tracking code to load and since it is at the start of the page, it will give an empty page to the visitor which is struggling to load the contents!Is that a good thing you would like to see as a visitor to a web page? I am sure you will be impatient.On the other side if you have the code at the bottom of the page, the visitor is provided with the page content before the web analytics code is being loaded. So this placement – at the bottom of the page evolved from the usability perspective.Disadvantage: As mentioned, the web analytics data capture begins after the page is loaded. This gives us the risk of the visitor closing the browser or navigating to another page by clicking a link. Result: web analytics data isn’t captured for that particular page / visitor.
Top of the page, inside element: You have the potential risk of having a slow loading web page in this kind of implementation, but here you tend to get a bit more web analytics data being captured compared to the method mentioned above. User sees the page after it is been tracked. This placement is widely used in eCommerce shopping carts and checkout pages. If you are using Google Analytics and capturing eCommerece transactions, you must use this placement to make proper calls to eCommerce tracking methods of GA.

Inside

of a page: This is not a widely used method, but can be handy in certain situations. Imagine you have another set of JavaScript code which accidentally has a variable name or a function name same as Google Analytics which you are using as the web analytics tool. Here you will have a conflict and either one or both of the code snippets malfunction. If you place the web analytics tracking code inside and above rest of the JS, you might eliminate this risk. Please Note: This particular method is suggested for Google Analytics. If you are using other web analytic solutions, please contact your vendor before using this method!

I would like to repeat, the most widely used JS tracking code placement is at the bottom of the page, just above . Tracking all visitors isn’t important most of the time, but to make the website function well to the visitor is the first and foremost criteria – which need not be forgotten!

Mobile Device specific reports in Google Analytics using Advanced Segments

seolion — Wed, 26 Aug 2009 10:49:08 +0000

I have gone through few interesting blog posts on tracking mobile devices using Google Analytics.

Basically I hve seen three methods:

Using a duplicate profile and using an include filter, to include traffic from mobile device operating systems.
Using the advanced segment feature of Google Analytics, creating an advanced segment of mobile device operating systems.
Using the advanced segment feature of Google Analytics, creating an advanced segment of mobile device screen resolutions.

Of the three, I feel the last two are better, considering the fact that we need not switch profiles to see mobile reports and reports containing all reports, you are saving a duplicate profile. Remember there is an upper limit for the number of profiles you can have in an account.

I have created these particular segments and would like to share it with you.

You can add this Advanced segment for mobile device traffic in Google Analytics based on Operating systems by clicking the preceding link. During installation of this segment on your account, you can actually verify how I have created this segment. I have actually used the condition ‘containing’ specific words in the OS names than actually giving the exact names.

Thanks to Bryson Meunier for his advanced segment using screen resolutions, read the post at http://www.brysonmeunier.com/mobile-analytics-with-google-analytics/ . I have modified his segment to include devices with higher screen resolutions and you can download Advanced segment for mobile devices on Google Analytics based on Screen Sizes by clicking the preceding link.

Test both the segments above and contribute your ideas to improve them and share with everyone. You can contribute to this initiative by suggesting any OS which is missing in list and by testing this segment on your account, or a device’s screen resolution which isn’t covered in the regex.

Post your thoughts!

I’m now a Google Analytics Qualified Individual with 100% score!

seolion — Fri, 07 Aug 2009 17:20:13 +0000

Google introduced test and certification for Google Analytics last March, since then It was running in my mind to complete it. I was preparing for my Implementation certification in Omniture SiteCatalyst, so thought going at GAIQ will be something good to start with.

Anyways Google Analytics and Omniture SiteCatalyst are entirely different web analytic tools, but getting certified in both is certainly a value addition to my resume, I have now cleared GAIQ with a maximum score of 100%

You can view my GAIQ test record at the official test site. Well, that shows my real name! (SEOLion is my nickname, please read my about page to know more about me)

Here is my certificate from Google Analytics which is valid for next 18 months!

Well, getting a high score in GAIQ may not be tough, I am sure not many get the magical number of 100%, luckily I have it!

GAIQ Preparation

Google has provided a comprehensive study material to prepare for GAIQ (Google Analytics Individual Qualification) at Google Conversion university. One should have gone through the material completely and understood key points discussed in order to tackle some of the tricky questions at the test.

The primary point to start with is a Google Analytics account. Since its free, you can try implementing on your website / blog. This I would say must to understand how Google Analytics captures data. After implementing GA code on to the site, you should have validated and seen that all pages are being tracked and shown on the report. This is the basic implementation. Now comes cross-domain tracking. This scenario is similar to tracking on a third party shopping cart without breaking the user session. Advanced implementation includes event tracking. I have experimented event tracking my selves, hence tackling questions related to that wasn’t tough for me.

Virtual page views is another important thing to know in Google Analytics, fortunately I had enough hands on on these while setting up goal tracking on an Ajax based site.

Filters, Goal conversion tracking related questions can be tackled if you have some prior experience working with these or you have gone through the material in detail and have done some extra research on these area.

Regular expressions is one of the tough area in the test, but one can make use of the regular expression builder to answer the questions.

I never had hands on on e-commerce tracking, but understanding the study material was sufficient to answer questions related to that.

Profiles in Google analytics was something new for me, but again the tutorial made it very easy to grasp it. I have duplicate profiles already put in action after learning them.

So overall my preparation lasted for a couple of weeks, could have been lesser if full attention was on this.

I would suggest one to take notes while going through the material. Having your account side by side can help you understand better. Since the test isn’t on an exam hall, you are free to have your account open in another browser, along with tools like regex builder. Going through the voice tutorial isn’t a very good option in between your test, your notes can get very handy if you have taken them enough.

Good luck to the ones about to take up GAIQ. Please let me know your queries, I am more happy to assist you.

What next?

Feel free to contact me if you feel I can be consulted for your project involving Google Analytics

Do you wish to follow my tweets in Twitter? Find my Twitter profile link in the contact page.

Cookie basics

seolion — Thu, 27 Nov 2008 14:18:06 +0000

What is a Cookie?

Cookies, HTTP Cookies, Web Cookies or Tracking Cookies are small pieces of data stored by a website on client system. This data is sent back to the web server which initially set the cookie for any additional hit to the web server.

What are the uses of cookies ?

Cookies are majorly used for user authentication and carrying the active session across different page visits by a visitor in a visit session.

Say user A has a login at xyz.com. There is a login page where A enters his login details and goes ahead browsing pages in XYZ.com. Now cookies can be a method by which xyz.com distinguish user A as the same person visiting different pages at xyz.com.

Web analytic tools use cookies to calculate unique visitors.

How a cookie is set ?

Cookies can be set by javaScript or any server sided script.

JavaScript cookie example

Setting a cookie
document.cookie = “COOKIE NAME”+”=”+”COOKIE VALUE”+ “;expires=”+”EXPIRY DATE”+”;domain=”+”DOMAIN”;

COOKIE NAME is a string as a name for cookie, Value is a piece of data, EXPIRY DATE is created by using Date.toGMTString() method of javascript. Expiry value and domain are optional.

Reading a cookie

Reading a cookie is not straight forward. All the cookies set for a domain are available in document.cookie variable as a string. We need to have utility functions to read a specific cookie.

PHP Example

Setting cookie
bool setcookie ( string $name [, string $value [, int $expire [, string $path [, string $domain [, bool $secure [, bool $httponly ]]]]]] )

Reading a cookie

The cookie is directly readable from the $_COOKIE global variable as $_COOKIE[$name]

Same origin policy related to browsers and cookies

In simple terms the explanation of same origin policy would be, if xyz.com sets cookie, only javascript code in a page at xyz.com or a server sided script at xyz.com can only read the cookie. This is a browser policy. If www.xyz.com sets the cookie, the same way only www.xyz.com page js code / server sided script can only read cookie.

Same cookie over all subdomains

If you would like to share the same cookie data across all subdomains of a site, say xyz.com, the domain parameter in a setCookie function should be .xyz.com (dot at the start of the base domain name. No www’s in it !!)

References

HTTP Cookie on WikiPedia

How web beacons perform web analytics data collection

seolion — Mon, 22 Sep 2008 10:07:19 +0000

How does web beacons work? if you have asked the question any time and would like to know still, read on !

If you are looking forward to know how does an analytics tag work or what is a web beacon, this post might help you.
A web beacon is usually a 1X1 pixel transparent image which gets embedded into a web page or an email message which acts as a way to exchange data between the client and the web analytics data collection server.

Lets see how web beacons really work and transfer data to the server.

Web beacons through javascript tags

The usual web analytics data collection method being used is by means of javascript tags. Each web page which need to be analyzed or tracked contains few lines of javascript code. Following is an example of statcounter code (which is taken out of this page itself, as I use statcounter as one of the web analytics tool. Its free!!)


<script type="text/javascript">

script>

<script type="text/javascript" src="http://www.statcounter.com/counter/counter_xhtml.js">script>
<noscript>

noscript>

Let me explain what this code does.

In first 4 lines of this code, few statcounter account related variables are being initialized (such as sc_project, sc_invisible etc). The last part of the code is inclusion of an external js file from the statcounter.

The code from statcounter does many things. For example it finds out the referring URL to the current page you are viewing, the page’s URL, title of the page, your browser’s resolution, your IP address etc etc).

Now comes the web beacon. The web analytics data is collected by the script from statcounter. And at the end of this initial process of collecting information, the data need to be passed to the statcounter server. This is achieved by use of a web beacon or image plug. An image request URL is formatted appending all the variables and values which need to be passed. Javascript outputs an tag with this URL as the src attribute. When the browser renders this image, the script at statcounter.com collects the data present in the query string and sends out the 1X1 transparent image as the response (it was a request for an image – remember that).

Web beacons as part of a mobile specific web page

You would have seen a