Understanding how users move through your system is critical for decision-making. Whether it's tracking customer journeys, product flows, or data pipelines. Traditional charts often fall short in showing relationships between stages. This is where Sankey diagrams shine.
In this post, we’ll learn how to build a Sankey chart in Apache Superset using a simple website user journey dataset.
Demo Dataset (CSV)
We’ll use this dataset representing user flow.
websiteTraffic.csv
traffic_source,page_category,action,user_count Google,Home,Signup,120 Google,Home,Browse,300 Google,Product,Add to Cart,180 Google,Product,Purchase,90 Facebook,Home,Signup,80 Facebook,Home,Browse,220 Facebook,Product,Add to Cart,140 Facebook,Product,Purchase,60 Direct,Home,Browse,200 Direct,Product,Add to Cart,160 Direct,Product,Purchase,110 Email,Landing Page,Signup,150 Email,Landing Page,Browse,100 Email,Product,Purchase,70 Referral,Home,Browse,130 Referral,Product,Add to Cart,90 Referral,Product,Purchase,40
Follow below step-by-step procedure to build Sankey diagram.
Step 1: Load Dataset.
Navigate to Data -> Upload CSV to database
Select the csv file, Database and Schema, give the table name as web-site-traffic.
Click on Upload button.
Reload the Dataset listing page, you can observe that the ‘web-site-traffic’ dataset is created.
Step 2: Create Sankey Chart.
Navigate to Charts listing page and click on Create Chart (+ Chart+ button.
Choose the dataset as ‘web-site-traffic’, select ‘Sankey Chart’ and click on ‘Create new chart’ button.
You will be taken to Sankey chart configuration page.
Configure
· Source → traffic_source
· Target → page_category
· Metric → SUM(user_count)
Click on ‘Create chart’ button, you can see that Sankey diagram created like below.
Why Sankey Over Other Charts?
Unlike bar or pie charts, Sankey diagrams:
· Show relationships between stages
· Highlight flow magnitude visually
· Make complex journeys easy to understand
Previous Next Home
No comments:
Post a Comment