Conversion Analysis Using Funnel Widget
Funnel analysis includes a series of events that lead towards a defined goal. For example, tracking user activity in a mobile app, a sale on an eCommerce website, online advertisements, and eventual purchase.
The funnel widget visualizes use cases such as:
- Retail application checkout drop-off rates from the point of adding items to the shopping cart to the actual purchase
- Business Journey drop-off rates from the point of applying for a loan to approving the loan
- Home mortgage loan approval drop-off rates from submitted application to approved applications
- Impact from performance issues by total sales, customer segment, location, and purchased products are gauged in a funnel with other health rule widgets
In this custom dashboard, the impact from performance issues by total sales, customer segment, location, and purchased products are gauged in a funnel with other health rule widgets.
Funnels return the number of unique actors that successfully (or unsuccessfully) make it through a series of steps. Actors could mean users, devices, or any other identifiers that are meaningful to you. A funnel analysis reveals where a given flow loses the most actors. This helps identify areas for improvement, as well as the overall health of your business.
The first step in the funnel is the anchor step, and always represents 100% of the results. The filter criteria for the first step creates a set of unique values, and every subsequent step results in a subset of the values from the previous step and renders the corresponding drop-offs. The second and later steps describe the number of events that match the criteria for each step towards the goal. This number is typically less than 100%, but could be 100% if every event matches the subsequent steps.
Build a Funnel Widget
- From the Analytics Search page, click Add > Drag and DropSearch. Select the Visualization tab.
- Click Add Widget > Funnel Widget Builder.
-
In the Funnel Definitionpane, click the down arrow to select the unique field representing the items that you are counting. The fields depend on your application. There is a limit of 50,000 unique values for the first step. If you encounter this limit, you can try decreasing the global time range and look at fewer values at a time.
Note: Filtering on fields with null and empty values may cause more records in subsequent levels. - Label the first step and continue to add steps as needed for your application process.
- Continue to add and label your steps.
- After you have added all the necessary steps, click Add & Close Widget Builder to save the completed funnel widget. If you want to cancel editing and revert to the previous version, click Close Widget Builder.If the funnel steps need a rearrangement, you can simply reorder them by simply dragging and dropping each step into appropriate positions.
- Once you finish building your funnel, remember to name and save your search result.
- To add the finished funnel widget to a custom dashboard, click Actions > Add to Dashboard.
Use the Funnel to Troubleshoot
The finished funnel widget for the above example shows the drop-offs (Abandoners)in unique loan IDs from submission to approval. When creating the funnel, you can also select Show Health in the properties pane to display a health overlay that correlates application health data with the business conversion data. This helps to visualize performance for each step of the funnel widget.
The health overlay is supported for these events types with a user experience value:
- Transaction Events –
userExperience - Browser Records –
pageexperience - Mobile Snapshots (Network Requests) –
networkrequestexperience - Web Session Records (Browser Sessions) –
experience - Mobile Session –
networkrequest.networkrequestexperience
For enhanced troubleshooting, you can drill down to event data from any of the funnel steps by double-clicking the numbers on conversions and abandoners or by using Health Overlay and clicking on each health segment.