Feature View and visualization improvements

Feature View

After spending a few months in our beta testing program we are happy to announce the availability of our new Feature View report. The Feature View report takes the existing Feature Usage report and allows for customization and aggregation. It allows you to pick and choose from the full range of tracked features and create custom views of these data, selecting only the features that matter to you and optionally aggregating features to form exactly the data visualization you need.

The Feature View is available as a new report item in the menu. Select the item and create your first view by selecting Create new view.

A dialog appears which allows you to customize a new view. You can specify a custom title for the view as well as a more elaborate description. To select the features to include in the view select Add new feature at the bottom of the dialog. This will bring up the feature selection dialog which allows you to select the features to include.

You can use the category selection to narrow down to the specific feature you want to examine and then pick out the features to include. Note that you canmultiselect features and either add the features as individual data series in the view or as an aggregated single data series. Aggregated data series simple summarize the individual feature usages into a new data series.

You can continue to add new data series by adding features as well as modify the existing data series. You can freely name the individual data series as you please if the automatically generated names does not work for you.

Once you are satisfied with the feature view you just click OK and the data view is populated with the data series you have assigned. This data can be accessed and viewed using the tools you are accustomed to such as viewing in data grid, exporting to excel or using charting options.

Controlling feature visibility and naming

When working with features over time you may need to remove data that is no longer relevant or rename internal feature names to something more human readable. This can be done using the Configure item in the menu for Feature Usages as well as Feature Timing and Feature Value data. When selecting theConfigure item two new items becomes available:

The Hide this item checkbox can be checked and thereby completely hiding the section from view. This can help slim down the view and hide sections that was perhaps added as part of initial testing or integration spikes. By clicking the Edit individual feature names link, you'll be presented view a dialog that allows you to control the names and visibility of the individual features in the section. If individual features are hidden, they no longer show up as part of the visualized data.

Note that you can show hidden features again by going back to the dialog and setting the feature to be displayed again. Similarly you can temporarily show hidden sections again by clicking the link at the bottom and select the view to display again. If you want to permanently show the section again, you should go into theConfigure menu for the section and uncheck the Hide this item checkbox.

Percentage visualization

Sometimes your data is harder to digest in its raw form. Converting and comparing numbers in your head is tiresome. We've easied the burden by allowing the data to be displayed in a percentage view instead of a showing the raw numbers. For instance, consider the following data visualizing the version adaption over time:

You can easily see the general distribution for the full period in the pie chart to the right and the general flow in the chart for the selected date range. But what if you need to speak of the general uptake of a new version in percentage over time? We added a new chart option called Percentage of total usages that will change the data visualization from the raw numbers to percentage of the daily usage count.

This makes it easier to compare data flows over time as well as easier to relate data to the overall usage to get a feel for the overall significance of the data. With the new chart option turned on the chart from above becomes:

Various improvements

A number of smaller improvements has also been added to the service, including improvements for exception handling and better supporting the integration and testing phase of your projects.

We are continuously taking customer feedback in order to improve our service in directions that makes sense to you. Keep the feedback coming and we'll use it to guide or development.