Archive for July, 2008

Jul 25
2008

No Data from One or More Sources in your Analytics Reports?

web analytics  By: Allaedin Ezzedin No Comments »

Have you ever experienced a call transfer from one department to another while you were calling to troubleshoot a problem in your machine or report a credit card fraud?

This annoying experience is similar to what we experience in web analytics when we deal with URL redirects. You request one URL and end up at another! Misconfigured URL redirects can cause data loss such as not seeing any data from one or more sources in your web analytics reports

It is common practice in many sites, especially e-commerce sites, to use redirect pages to track campaign performance. The problem is that the redirect usually removes extra parameters from URLs – parameters which are necessary for proper tracking in Google AdWords and Google Analytics.

In order to identify a visitor as a paid visitor in Google Analytics, AdWords auto-tagging adds a parameter to the end of any AdWords destination URL. This parameter is called gclid.

In a normal situation with no URL redirects, when people click on a paid ad with a destination URL such as http://www.mydomain.com/landingpage.aspx?id=54

the URL that they are supposed to go to might end up looking like this http://www.mydomain.com/landingpage.aspx?id=54&gclid=a1b2c3d4e5f6g7h8i9

If there was a poorly configured redirect at this point, the visitor might end up at this URL: http://www.mydomain.com/landingpage54.aspx *

* Note that the gclid parameter is gone and Google Analytics will consider the click as an organic visit and not as a paid visit.

Suggested solutions to this problem:

  1. Ask your webmaster to configure the redirect page to pass any parameters to the final URL. This is will allow you to maintain your internal redirects and properly segment your visitors.
  2. Tag your destination URLs manually:
    • utm_source = google
    • utm_medium = cpc
    • utm_term = your AdWords keyword (ex. e-nor blog)
    • utm_campaign = your AdWords campaign (ex. blog campaign)

    The final URL will look like this now: http://www.mydomain.com/landingpage54.aspx?utm_source=google&utm_medium=cpc&utm_term=e-nor%2Bblog&utm_campaign=blog%2BCampaign

  3. The best solution: do not use redirects at all!  My colleague at E-Nor, senior web analyst Rehan Asif, suggests not to use redirect at all. He says, “configure Google AdWords to send visitors straight to the landing page. Add whatever internal tracking scripts you want to us on the actual landing page.” (ex. http://www.mydomain.com/landingpage54.aspx)One of the reasons we usually suggest avoiding URL redirects is because of our concern about the keyword Quality Score. Page load time and number of redirects are an important factor in determining the Quality Score. Redirects could be slow at times which would lead to long page load times which would lead to low Quality Scores.

Thank you,

Technorati Tags: , ,

Tags: , ,
Posted in web analytics | No Comments »

Jul 16
2008

Tracking E-mail Clicks across Multiple Pages

web analytics  By: Allaedin Ezzedin 2 Comments »

We often have an email address that is scattered all over our website and we not only aim to track the number of clicks on this email but rather also track the pages that this email is posted at

In this blog we will go over very simple steps of how to achieve the above goal

1. Add the following java script code to the main header or to the template file default.aspx

2. Add a second set of JavaScript code inside every email tag

To view report:

  • Log into your Google analytics account (www.google.com/analytics)
  • Go to the Content section -> Top Content
  • Filter the list of URL’s to only addresses contain “email_click”

Also we can track the email clicks using the Goal section!

  • Go to: Profile Settings -> Conversion Goals and Funnel
  • Create a goal as described below

  • To view the data go to: Goals -> Goal Verification

Technorati Tags: ,

Tags: ,
Posted in web analytics | 2 Comments »

Jul 15
2008

To Web Analytics and Beyond

web analytics  By: Shiraz Asif No Comments »

Hello world! I’ve always wanted to say that :)   I’m new to the blogosphere – I’ve posted on this blog before, but it’s been an eternity.  I’m hoping to be much more consistent going forward.  This post and the upcoming posts you’ll see from me will be focused on communicating what I’m learning about Web Analytics.  I’ve been in Creative Marketing now for several years – I have a strong background in Web Usability and Marketing Consulting.  You could say I’ve always been analyzing, just not with the tools/concepts in Web Analytics.  Although our company is a Google Authorized Analytics Consultant (GAAC), I’ve only been involved in a limited capacity in Analytics. That’s changing now :)   I attended an event held by the SVAMA 10 days ago and figured my notes about the event would be the best way to get the ball rolling.  Being new to Web Analytics, it’s a challenging subject to start writing about, but I figured there’s no better way to learn than to write about what I’m learning.  This helps reinforce what I’m learning in my own mind, and also opens the door to feedback from readers.  As I learn new concepts and techniques each week, I’ll blog about them.  I’m pretty excited about it, and hope this is useful and beneficial.

It’s 1:00 AM as I’m writing this and it’s been a super long day at work.  My 9 month old baby has been having tummy trouble the last few days so I’m feeling a little sleep deprived. In any case, I’ve had my coffee and everyone’s asleep, so it’s quiet and I can focus on writing :)   I have this new found love of woodworking and wanted to spend time in the garage but if I started cutting wood with my electric saw at this hour (no man is complete without an electric saw), I’d wake up the neighbors.  Besides it’s too late to finish that workbench I started building this weekend…I’ll get back to that next weekend I suppose.

To Web Analytics and beyond…

I recently attended a session held by the SVAMA at the Googleplex, with guest speaker and Analytics  Evangelist, Avinash Kaushik.  Before I start off, I want to say that Google food IS all it’s cracked up to be… my colleagues and I thoroughly enjoyed breakfast :)

Avinash is author of the book: “Web Analytics:  An Hour a Day.” He’s a wonderfully articulate, inspiring speaker.  As a newbie to analytics, I learned more from this session than what I had previously heard or read about.  We have several Analytics geeks here at the office and I hate getting left out of the conversation when they start talking in their ego-maniacal uber cool geek-speak.  Well, after this session, I can at least keep up with them, even if it  hurts a little.  :)

The title of the session was Actionable Web Analytics.  Avinash began by talking about data and how there is an overwhelming abundance of data available to companies.   There’s so many tools out there that result in data overload. He posed the question:  Does all this data provide us the insight needed for a successful site?

He emphasized 6 points which I’ve listed below.

1)  Actionability is an Attitude
The concept of actionability helps identify underlying issues on websites really fast.  Without much knowledge about the site or the subject matter, you can easily find Actionable Insights by looking at key metrics.

Bounce rate (To quote Avinash “I came, I puked, I left”) is a great way to measure how effective a page is performing.  Generally speaking, the lower the bounce rate, the better.  This has to be taken with a grain of salt though, since some pages might be designed to have a high bounce rate (ie. blogs, contact us pages, etc.)

Another point emphasized was about site home pages.  Too much emphasis is placed on the home page.  I’ve participated in hundreds of web site design cycles, and I can wholeheartedly agree that far too much emphasis is placed on the home page.  The concept of a home page assumes that the user will always walk through the front door into your home.  That simply isn’t the case.  We don’t live in a world where users have to enter your site through the front door.  The search engine holds the keys and the blueprints to your site.  Whatever the search engine determines is the home page, essentially becomes the home page of your site.  If your user searches for a keyword that results in the search engine displaying your “About Us” page as a result, than that’s effectively the home page for that user, for that visit.  So basically every page is a home page.

Armed with the above knowledge, you could now look at the top entry points of your site and easily determine what is working well and what isn’t.  This metric gives you information about your users (ie. what they are looking for) and also tells you what they are doing on your site, so you can focus on optimizing pages that need it.

2)  Give your Data Context
Avinash had this neat little acronym he called PALM – People Against Lonely Metrics.  I laughed, but then realized what it really means.  Data needs context.  Without context, the data is meaningless and definitely not actionable.  Data is relative and needs comparison to have meaning.  A report showing a month’s data isn’t as insightful and actionable as a report showing a comparison of the current month to the last month and to previous months.  This latter approach facilitates actionability whereas the previous approach would simply show a view of the data inside a bubble.  In highlighting metrics, focus on goal conversion and outcomes – these are what’s important to your client, the rest is just extra information.  Use tools such as compete.com or Fireclick to compare results to your competitors.

3)  Say No to Data Pukes
This section really opened my eyes because it cemented in my mind what Analytics is about.  It’s not about throwing a bunch of data and numbers to your customers.  Showing them top 10 pages or top 10 keywords is useful in the beginning stages but this list becomes stagnant very quickly.  How often does it really change?  All the little tweaks being done on the site – the ones that really matter – typically won’t result in a significant change in the “top 10″.  What’s important to highlight is what’s changing on the site, such as the top 10 rising pages, or the top 10 falling pages.  Again this gives data some context and is actionable.

4) Segment Like Crazy
I come from the Usability world, so when Avinash mentioned his next point about Personas, it connected several dots for me.  Personas!  Personas!  Personas!  it’s all about personas!  In order to understand your users, you have to know who they are where they are coming from and what they are doing on your site.  In analytics, data can be segmented by visitor source to identify where traffic originated from (ie. Organic search engine traffic, Paid campaigns, direct, referring sites, etc).  Visitor paths, etc. can identify what your users are doing on your site and give you insights on what you can do to make their experience more beneficial.

5)  Ask Your Customers
Surveys are a great little tool, revealing tidbits of information that is very valuable.  Ask your customers “Why are you here?” and “Were you able to find what you were looking for?” Two simple questions – but they provide feedback that will help you transform any site from a mundane to a rich experience, identifying what Avinash referred to as “segments of discontent” – another actionable insight.

6) Automating Actionability

Avinash has these creative little acronyms to illustrate his points – HiPPO is another gem of an acronym.  He said that websites are typically designed by HiPPO’s -  HiPPO stands for Highest Paid Person’s Opinion.  This brought back so many memories for me because I’ve facilitated so many creative designs and this is so true.  The only way to really tell which banner works best or which font style works best or which color works best, is to let your users decide.  Huh? but how can they participate in the creative review cycle?  Easy…  By doing A/B testing or multivariate testing, your users literally tell you what works and what doesn’t work.  Actionable Insights….

If there’s one thing I got out of this presentation by Avinash, it was that Analytics is Qualitative.  What????  Qualitative????  I almost wanted to press the rewind button to make sure I heard that right.  Wouldn’t it be great if life was a Tivo? :)   Qualitative?  Analytics is all about numbers right?  It’s about the data, right?

WRONG

This was the biggest revelation of the day for me.  All this time I was under the impression that Analytics was all about numbers and data.  It was purely quantitative in my mind.  In reality it’s just the opposite.  The numbers are only indicators of what is essentially qualitative data.  If you focus on the numbers, you’ll get lost.  If you focus on what the numbers represent (ie. user behavior), therein lies the heart of analytics.  Once I got this point down (and it took a while to sink in), it changed my perspective on what analytics really is and how it all fits together.

The last thing that he mentioned that I really liked was that companies should spend 10% on the tool, and 90% on the people.  Again this ties back into analytics being a qualitative discipline – it’s not just number crunching, we’re talking about user behavior.

Well time to go catch some zzzzzz’s now.  Look for more entries from me as I continue my learning curve in Analytics.  Looking forward to hearing feedback, and feel free to send me any tips about woodworking as well :)

Technorati Tags: , ,

Tags: , ,
Posted in web analytics | No Comments »