Widget Analytics – Measuring the widgets in the wild

Helping web analysts navigate the measurement and tracking of widgets.

  • Share this blog

    Bookmark and Share
  • December 2007
    M T W T F S S
     12
    3456789
    10111213141516
    17181920212223
    24252627282930
    31  
  • Meta

  • WAA

    WAA Member

Archive for December 18th, 2007

Measuring widgets – what skills does it take?

Posted by widgetgirl on December 18, 2007

So what skills does it take to not just measure widgets, but analyze their behavior? If you are a traditional web analyst like myself, you at least have a good foundation for understanding data collection, parsing, filtering and reporting. Measuring widgets outside of a somewhat controlled environment (your own web site or sites that is) is not as easy as it sounds.  The most important aspect is expanding your breadth of knowledge across the internet space.  Understanding the tools that are used, how sites host content, how data can or cannot be collected is a mandatory requirement for getting started.  Putting the pieces together to answer business questions and optimize how visitors engage with your widget and where it travels is the next step.

The critical skills for a web analyst:

  1. A strong understanding of web analytics – what I mean by this is that you understand the three core metrics of unique visitors, sessions, and page views for starters. You understand how to analyze incoming traffic to your web site and where people go through path analysis.
  2. Social networking smarts – anyone analyzing widgets and how they traverse the social networking space needs to be out there on the social networks. Web analysts should set up accounts on Facebook, MySpace and perhaps Friendster or MyYearbook. It depends on the space that your company is advertising or marketing within (but Facebook and MySpace are prerequisites for now).
  3. Start page smarts – a subcomponent of social networking is the ability to grab content and build your own start page on the internet. The most common ones are iGoogle, Netvibes and Pageflakes, but there are several out there to explore.
  4. Version analysis – browsers, operating systems and flash versions change with the wind. Understanding the changes in default security settings and capabilities is key to analyzing data out in the wild. Wild widget data is dependent on third party cookies and being first, second or third to load on a page. There is a lot of competition.
  5. “Environmentally aware” – Analyzing widgets in the wild is far more complex than keeping up with the changes that are going on within your own site. There are a lot of methods for keeping up with the environment, but the main ones that I would recommend would be following sites such as TechCrunch, Mashable and Media Post. It goes beyond that though – you need to have a critical eye to those factors that will effect your analytics.
  6. Troubleshooting skills – Finding the needle in the haystack is an art. When it comes to widget analytics, you can’t always be “right there” to recreate how a page loaded, which tag was called first, second, third, fourth……you catch my drift? I have spent the last few weeks trolling through use cases to understand our customers Google Analytics tags, Omniture tags and so forth. Some of these tags are firing from within a flash content file that is within a widget container that is within a page. Combining art and science is sometimes the only way to understanding and documenting the corner cases in the environment.
  7. Methodical and tenacious – you must be patient to do widget analytics. Most people would never call me patient, but believe it or not I have mellowed quite a bit in the past few years 😉 Ok, back to #7 here – you need to be able to work through the data in a very linear fashion. You can evaluate and analyze the anomalies, but you also need to follow the course. A widget starts as a seed placement and then grows from there. Analyzing it’s growth cycle should be done from the seed out to the branches. Optimizing that cycle to understand when and where your widget is the most “alive” and when it is starting to “die” is key in taking widget analytics to the next level.

Widget analytics is an exciting opportunity for web analysts to grow their skills off-domain. There are a host of new challenges in learning the environments where widgets travel and grow – some you can control and some you cannot. You need to be able to understand the difference and most importantly explain those differences to your internal and external clients.

And my favorite widget of the week…..this widget is promoting the new movie “Cloverfield” by Paramount. I just sat and watched the 5 minute trailer that plays in this widget and it is totally freaky. The cool thing that Paramount did with this widget is to make it a contest to see who can spread it the most. So analytically, what does that mean? It works like this – a visitor who views the widget must click on the widget and register their widget placement with their email address. Each new placement that is created from that particular placement will be counted towards the individual registered user. The registered user with the most new placements wins! The prize is a private screening of the movie in their home town – pretty cool. I don’t qualify of course as I work for Clearspring….so go register and check it out. More on this widget in my next post probably….there are a lot of cool widget analytics angles to hit on this one.

Cheers! Jodi

Cloverfield

Posted in widget analytics | Tagged: , , , | 2 Comments »