Learn how engagement in Customer Journey Analytics provides insights about how often your product features are used versus the number of people using them.
Hi, this is Michelle Hagar, and I’m a technical marketing engineer for the product enablement team. In this video, I’ll show you the feature matrix engagement analysis in Adobe Product Analytics. The engagement view helps fuel investment decisions by understanding what my core power one time and questionable features are. I’m able to look at features by reach of active users and frequency of use, which helps me invest in high value features while also balancing this with driving adoption for underused features. The guided analysis user interface lets me formulate a question in the query rail on the left, and then answers with insights, a matrix chart and a table on the right. When Engagement view loads, it auto selects my top five events or features based on how recently and frequently those events are queried in my organization. I can adjust the events to add up to ten to show in the matrix. It’s beneficial to add as many as possible for broad comparative results. I’ll share this tip as I add my features. If your feature is countered by several events occurring, you can define a new event in the data view to represent that feature. Then in the engagement view, you can select that newly created event from the list. Let’s review the matrix. It’s divided into four quadrants. Spanning the y axis is the scale for frequency or average times per user daily. Spanning the x axis is the scale for reach or average percent of daily active users. The quadrants are formed by calculating the medians for daily active users, and the median times used by day for the group of selected events. The top left quadrant represents power features. Those are not widely adopted across your broader user base, except for a subset of highly engaged users. The bottom left quadrant represents low impact features, those that are neither widely adopted nor frequently used. The top right quadrant represents high impact features. They’re both widely adopted and frequently used. And last. The bottom right quadrant represents one time features. These are widely adopted, but they’re not frequently used. Further below is the data in table format. If you want to compare features across segments, you can add up to three segments in the query. Let’s compare paid subscribers with free subscribers in the chart. As you hover over a point. Its companion point across segments will also highlight, allowing you to quickly compare feature engagement across your key user groups to see how features are resonating on a more granular level. Let’s look at some of the settings offered, which enable you to glean deeper insights from this view. Starting with median overlays, the default standard setting shows the absolute value of usage and engagement. The normalized settings shows relative changes from each median. The top events overlay shows you how the selected features compare to the top 20 events, based on the same algorithm mentioned earlier. The recency and frequency of these events being queried in your organization. This helps point your selected features in the perspective of your broader product. Time comparison for this analysis is useful for seeing how your feature usage improves or declines. I’ll select previous period to compare the current 30 day period with the previous 30 days. In the matrix, the comparison data is added as points and are connected to the current period points with comments. This view can help you understand if the feature improvements you’re making over time are having an impact on usage and user behavior. Any good analysis should spur new questions and actions for you. Like with other guided analysis views. There are a couple actions I can take from the chart or table. First, I can create a segment of users to further dive into other views or workspace projects. I can also view usage trends for any selected feature. This option opens the Usage Trends view and a new tab with the selected feature applied in the query. This enables you to seamlessly move from an aggregated to trended view to ask and answer new questions from. This has been an overview of the feature matrix engagement view in Adobe Product Analytics. I hope this comes in handy the next time you’re making product investment decisions.
For more information, please visit the documentation.