What is a funnel chart?
A funnel chart is used to visualize data that is progressively reducing or a linear process that has inter-linked stages. When the data is sorted using this tool, it takes the shape of a funnel. Each stage of the chart is a % of its previous stage.
For example, an e-commerce business who is evaluating a shopping cart process can use a funnel chart to best visualize the customer journey.
A typical shopping cart funnel would look like the funnel below:
When to use a funnel chart
Funnel charts have several benefits. Some examples include revealing bottlenecks, providing an overview of the health of a process as well as identifying areas of improvement (or where your funnel is leaking).
While these charts have been historically used for evaluating sales conversion data, they now have a wide application given the emergence of the Internet of Things (IoT).
How to create a funnel chart
Here I’ll create a funnel chart using Chartio, a business intelligence software tool that simplifies data visualization and data analysis. You can follow along by logging in and using the Chartio Demo Source data set.
- Create a New Dashboard and click on Add Charts. This will take you to the Explore page which allows creating charts using Interactive mode or SQL mode. Choose Chartio Demo Source as the data set.
- Expand Visitors, drag & drop # Visitors ID to the Measures section. Rename the label as Visitors for clarity and click ‘Ok’ and ‘Run Query’. Change the default data visualization (number) selection on the right to a Table. (The 2nd column of the table should be numeric and non-negative to ensure that your data is organized accordingly. You can drag & drop to rearrange the columns in the interactive mode)
- Here you will want to combine multiple data sets. You can do that within Chartio using the Data Pipeline transformation step “Add Layer”. It will automatically add a join path.
- Rename Layer 1 as Visitors and add another Layer (Layer 2). We do this to create the second phase of the funnel chart.
- We repeat the process for Users. Expand Users, drag & drop # User ID to Measures. Rename the label as Users, click Ok and Run Query.
- Rename Layer 2 as Users and add Layer 3. Expand Subscriptions, drag & drop # Subscription ID to the Measures section. Rename it as Active Subscriptions. (We need to ensure that Non-Active/Cancelled subscriptions are filtered away at this point for the current layer).
- Drag & drop Cancelled Date to the Filters section and select is null from the drop-down list in the first tab. Click Ok and Run Query. Rename the Layer as Active Subscriptions.
- Add a fourth layer. Expand Subscriptions, drag & drop # Subscription ID to the Measures section. Rename it as Cancelled Subscriptions. Drag & drop Cancelled Date to the Filters section and select is not null from the drop-down list in the first tab. Click Ok and Run Query. Rename the Layer as Cancelled Subscriptions.
- Merge Layers using union under the Merge type drop-down list. Check the box for Add input names and Apply & Close. The data is now sorted in a descending format which accurately represents a funnel.
Best practices for using funnel charts
A funnel chart can be a great choice when the data is interlinked or sequential. It can also work effectively for evaluating sales conversion rates as mentioned before, or for assessing the progress of a click-through advertising or marketing campaign.
A funnel chart is typically used when the data in the first stage is larger than the rest and is visibly reducing until the last stage.