Widget Analytics – Measuring the widgets in the wild

Helping web analysts navigate the measurement and tracking of widgets.

  • Share this blog

    Bookmark and Share
  • August 2017
    M T W T F S S
    « Dec    
     123456
    78910111213
    14151617181920
    21222324252627
    28293031  
  • Meta

  • WAA

    WAA Member

    WAA

Posts Tagged ‘widgetwebexpo’

WidgetWebExpo – Widget Analytics

Posted by widgetgirl on June 19, 2008

My good friends Ivan Pope and Jim Sterne hosted the WidgetWebExpo conference this week in New York. I had the pleasure of co-presenting with a new friend, Albert Lai from Kontagent. My portion of the presentation focused on widget analytics and Albert’s focused on measuring Social Applications. This was one of the first conferences that I have ever been to that was 100% focused on widgets – an interesting venue when everyone in the audience is actually focused on widget analytics and measurement.

The topic of the panel was “The What, Why, Where and How of Widget Metrics.”  You can see me in action presenting the topic here on YouTube.  Got to love the Flip Video cams that people carry these days.

I started off with “what is a widget?”  I know that seems like such a silly question, but honestly  the question still comes up all of the time. I always start with the Wikipedia definition, but when it comes down to it, the easiest way to describe a widget is to think of it as a “mini web page” that has options for you to transport it from one web site to another. Widgets can be run on web page, desktops and even some mobile phones.

So once we tackled defining the space, we drilled down deeper into what makes a widget sharable.  I mean, what if you could just copy the source of a link and place it on your own site?  Does that make it a widget or does the piece of content have to come with a sharing mechanism in order to cross the bridge over to widgetdom?  I don’t claim to answer the answer to that question from a technical standpoint, but I think we all agree that the content needs to be “transportable”. The example in the slide below shows the visitor experience when grabbing a widget and placing it on iGoogle.  There are essentially a few basic steps:

  1. Click on the “grab” button (also called share or add button).
  2. Select the destination site to share the widget to or grab the embed code.
  3. Authenticate to the site that you wish to share the widget to.
  4. Enjoy widget!

Using a widget serving platform has multiple benefits – the platform builds all of the “bridges” to the share destinations (so you don’t have to) and makes it much easier for the user to grab the content.  Just exposing the embed code will be problematic.  Although there are some visitors who understand how to copy and paste embed code into their start page, blog or social network profile page, the majority of users don’t know how to do that.  The goal is to make it as easy as possible.

In prepping to transition the audience into a true metrics discussion, you have to tee up the fact that metrics should be the tool to measure how well you executed on your strategy.  One of my favorite quotes of all time is “Begin with the end in mind” from Stephen Covey’s book The Seven Habits of Highly Effective People. You need to understand who you are trying to target with your widget before you ever release it.  What is the purpose of building a widget – to build off domain audience, drive people back to your site, sell something or just build brand awareness?  Before you get set to measure, you need to know what success looks like?

Once you know who you are trying to reach, then you need to go seed your widget in spots where people will find it.  The goal is to make it as viral as possible by placing it on pages and in networks where it has the highest probability of being grabbed.  The “if we build it they will come mentality” is not going to work here.  You need to be methodical about putting your widget in spots where:

  • Visitors will know that it is a widget
  • Visitors will understand how to grab it
  • Visitors will be compelled to grab it (creativity obviously plays a part here, but so so does aligning the widget with content that is relevant)

There were three sets of analytics that I discussed with the audience.  We have covered many of these here in other posts, but they categorize into three areas: Spread, Audience Extension and Engagement.  Spread analytics focuses in on the content itself – where is it being grabbed from and to which domains is it being installed? You’ll notice that I am starting to use the term “Installs” over “Placements” in my terminology.  Here at Clearspring we are starting to standardize around the term as it aligns with how the industry is defining sharable content.  More to come on this in future weeks.

The three metrics I discussed were: Grabs, Installs and Active Installs.  Grabs being the number of times that a widget was “grabbed” from another widget.  Installs being the number times that the widget was successfully installed (we require it to be viewed at least once on the destination domain or desktop). Active Installs being the number of widget installs viewed at least once during a selected time span.

All three of these metrics are very unique and critical to measuring the success of a widget.  How viral is your widget (how many times has it been grabbed and how quickly is it spreading)?  Also critical is understanding where your widget is traveling to and whether or not that install base you are creating is sticking and growing or simply churning and being replaced over time.

Audience extension consists of derived metrics that help you understand two core business questions – how successful are you at generating content consumption off-domain and how well is your content spreading off-domain.  When I say “off-domain”, I mean not on your own web site or widget seeds.  These two metrifcs are intended to help quantitatively measure the reach of your widget.

The third category of metrics that I discussed were on engagement.  So how is someone “engaging” with my widget?  Seriously, I hate that word.  Everyone uses it and it is so incredibly silly.  That aside, my take on engagement is that it is simply a word that acts as an umbrella across the metrics available to measure how a visitor interacts with your content.  In the widget space there is a lot that you can measure inside that “mini web page” that is portable.  On the Clearspring platform we measure interactions (mouse overs or clicks), clicks and time spent.  Those three metrics alone are used to derive additional metrics that can all be used as tools to measure widget activity and most importantly how visitors are interacting with your content and brand.

So last, but not least, I wrapped up the presentation discussing how to optimize your widget campaign.  It starts with working with a widget platform that uses sharing tools that are easy for the visitor to use.  You want a platform that has broad reach in the destinations that they work with so that you can reach as wide of an audience as possible – and of course you want a platform that will provide all of the analytics so that actually can tune and optimize your widget.

A/B testing with your creative is also a component for understanding what your visitors will respond to.  The messaging in the grab button, the features of the content and the freshness of the content will all contribute to how successful your widget is.

Promote your widget through ad campaigns and affiliate networks – seed, seed, seed!  The more places you seed the widget (within reason and business smarts of course) will help fuel its virality and distribution.  The entire process is a cycle (Lather, rinse, repeat).  Modify, measure, tune/optimize.  Effectively working through this cycle and making it part of the DNA in your organization will help you create widgets that are viral and effective at meeting your business goals.

Posted in widget analytics | Tagged: , | Leave a Comment »