Funnel Charts Overview:
Funnel charts are a popular data visualization tool that are frequently used to represent the stages of a process or the conversion rates of a business funnel. Funnel charts help users to more easily identify areas of improvement in processes and make data-driven decisions to optimize results.
Some use cases of funnel charts include sales and marketing funnels to track the process of lead generation to sale completion or what we will be going through today: recruitment process funnels to track the number of applicants through their interview process down to the ones who accept their final offers.
For today’s example, we will be utilizing a dummy dataset following 900 applicants for three positions: Account Executive, Software Engineer, and HR Specialist. Each position has seven sequential “stages” of the recruiting process:
- Phone Screen
- Interview #1
- Interview #2
- Panel Interview
- Offered the position
- Accepted the position
All of this is reflected in a dataset in a wide format such that we have four columns:
- ApplicantName: the name of each applicant
- Position: the position the applicant is applying for
- Stage: which stage in the application process that each row represents
- Status: whether or not the applicant has successfully entered that stage (e.g. True for stage “Phone Screen” means that they engaged in a phone screening interview, while True for stage “Accepted” means they have accepted their offer).
Step-by-step process to building Funnel Chart
Step 1: Create a child element
From our base table we will want to create a child element by in the top-right of our data element clicking “Create Child Element” → Visualization → Funnel
Step 2: Assign Stage Value to Stage
Next, we will want to bring our column that represents each stage in our process (in this example, titled “Stage”) over to the Stage section for our visualization.
Step 3: Bring in Values
Next, we will want to quantify how many applicants we have at each stage of our interview process. An important note: this process may vary depending on how your data is structured. In this example, we have a row for every applicant for every stage of the application process, regardless of if they passed or not because we have our “Status” column which indicates whether or not that value should be counted via True/False. Because of this, we can use the CountIf([Status) which will only count the value if it is equal to True.
It is not uncommon for data to be structured such that rows are only added for each applicant who made it to the next stage of the application process. For example, if an applicant made it all the way from the Application stage to the Acceptance stage, they’d have 7 entries of their name, while an applicant who never made it past the Application stage would only have 1 entry. In cases like this, you would only count each row using Count().
Step 4: Let’s get formatting!
Now the fun part – formatting! While not an exhaustive list, I will cover a few of the formatting properties that are unique to funnel charts.
- Show Value: this will show the raw number of each stage, represented by the white number within each stage below.
- Show Stage Name: this will show the name of each stage, represented by the values on the left side of the chart below.
- Show Conversion Rate: this will show the percentages of each stage, represented by the values on the right side of the chart below. The two options are discussed in the next bullet.
- Conversion Rate Percentage Style: there are two options: “% of total” (in blue below) and (2) “% of prior” (in yellow below). Selecting “% of total” will take the number of applicants at each stage and divide it by the number of applicants of the first stage. Selecting “% of prior” will take the number of applicants at each stage and divide it by the number of applicants of the previous stage.
In conclusion, funnel charts provide a easy and useful way to visualize the different stages of a process and potentially identify areas where improvements can be made, as seen by our employee application example.