Widget Installation Base – Entrenching your brand
Posted by widgetgirl on April 29, 2008
Distributing your content in the form of widgets is like infiltrating the internet with your brand. As widgets are grabbed and placed on web sites, their staying power is where the impact of the widget really shines. I have written about this metric before – it is core to widget analytics. The Active Placements metric is critical in measuring how well you have spread the widget and whether or not your widget installation base is active, expanding or contracting.
When a widget is grabbed and placed, we measure this with two separate metrics:
- Grabs: the number of times that a widget was grabbed. This metric is cumulative in nature and is grouped by the source domain that spread the widget.
- Placements Created: the number of times that a widget placement was created and viewed at least once on a domain. The criteria of “Viewed at least once” is important because someone could grab the embed code for a widget (or use the built in sharing services of a widget platform), but never actually place the widget. When this happens, we call it a grab attempt, but not a successful widget placement.
To put this into context, if I grab a widget from Facebook.com and place it on iGoogle, this would count as one Grab from Facebook and one Placement Created on iGoogle. Each of these metrics are specific to the action of spreading the widget. However, once the widget has been placed on a web site, we then need to measure how long it stays on the user’s page. Widgets can be deleted and they can also go dormant where they just aren’t viewed anymore.
Once the content has been placed, measurement of the install base is the next step for understanding if your content is sticking to where it was placed or if it is simply being placed and then churning off. The metric to analyze the install base is as follows:
- Active Placements: The count of widget placements that have been viewed at least once during the specified time period.
Let’s imagine that we have the following scenario:
- A widget is grabbed 5,000 times resulting in 5,000 Placements Created.
- After one week we have 5,000 Active Placements for the week.
- During week two the widget is grabbed an additional 500 times resulting in 500 new Placements Created.
- After two weeks we have 4,600 Active Placements for the week.
Is this a good scenario or a bad one? It is actually not a good one from a longevity standpoint. Although having an install base of 5,500 widget placements is good, the fact that the Active Placements metric is declining while new placements continue to be created indicates that the install base is churning faster than the spreading of the widget can replace it.
There is always going to be churn to your install base – so goes the life of the widget. One needs to watch this metric and react when there is a negative trend. Refreshing the content of the widget and/or finding new web sites or locations within a site to seed the widget can help boost the stickiness of the widget and how long the widget lives on a user’s page.
iVillage has a widget that is one of the most popular on our platform. This widget has been placed over 850k times…read that…850K times. The widget has slowly made its way to the top of the Friendster Widget Gallery and has gained popularity on that particular network. The stats for the widget are available on the widget’s home page. Active Placements aren’t reported here, but I am sure that the widget creator is watching that metric and optimizing based on it’s trend 🙂