Clustering in Tableau - Cell-by-Cell Formula Trace

Sample Data

Sample customer data with Annual Income and Spending Score used for clustering analysis.

Cell	Value
A1	CustomerID
B1	Annual Income
C1	Spending Score
A2	C001
B2	45000
C2	39
A3	C002
B3	54000
C3	81
A4	C003
B4	32000
C4	6
A5	C004
B5	76000
C5	77
A6	C005
B6	58000
C6	40

Formula Trace

CLUSTER([Annual Income], [Spending Score])

Step 1: Identify features: Annual Income and Spending Score for each customer

Step 2: Apply clustering algorithm (e.g., k-means) to group customers based on feature similarity

Step 3: Assign cluster labels to each customer

Cell Reference Map

    A           B               C
1 CustomerID  Annual Income  Spending Score
2   C001       45000           39
3   C002       54000           81
4   C003       32000            6
5   C004       76000           77
6   C005       58000           40

Features used: Columns B and C for clustering

The clustering formula uses Annual Income (B) and Spending Score (C) columns as input features.

Result

    A           B               C               D
1 CustomerID  Annual Income  Spending Score  Cluster
2   C001       45000           39             2
3   C002       54000           81             1
4   C003       32000            6             3
5   C004       76000           77             1
6   C005       58000           40             2

The final output adds a Cluster column showing the cluster number assigned to each customer.

Sheet Trace Quiz - 3 Questions

Which columns are used as features for clustering in this example?

A Annual Income and Spending Score

B CustomerID and Spending Score

C CustomerID and Annual Income

D Only Spending Score

Key Result

CLUSTER groups data points based on similarity of selected numeric features.