Market Simulation

Coke Zero Cherry – Cannibalization

This Market Simulation is designed to simulate the degree of same Brand Cannibalization if Coca Cola were to introduce Coke Zero Cherry.

Fictitious Scenario: Coke experiment by introducing a new Product Variation into their assortment. They develop “Coke Zero Cherry” and sell it in a can in bundles of 6×330 ml.

At first the sales look good. From a total base of about 10,000 monthly Cola Customers, the new product sells about half of the current sales of “Coke Zero” (“Coke Zero Cherry” sells 128 bundles per month versus 211 “Coke Zero” bundles).

But at second glance, the total Quantity, Revenue, and Profitability grow at a much smaller rate. Revenue is up from $96,125 per month to only $96,985 (about an $850 increase).

It turns out that over 90% of the new Coke Zero Cherry’s sales were due to Cannibalization. That is, over 90% of new Customers would have bought another Product from Coke but were induced away to buy “Coke Zero Cherry” (109 old Coke customers versus 19 Pepsi or other brand customers).

This Case Study provides a high-level overview of the workflow without detailed explanation. It assumes you are already somewhat familiar with KNIME and Market Simulation. If not, start by reviewing the Building Blocks and Community Nodes.

#1 Product Generator

The Product Generator Node will create a predicted Output Willingness To Pay (WTP) Matrix for the new Coke Zero Cherry Product. The part-worth values for each Feature were derived from the Feature Tuning Workflow for the Cola Market.

See also: CS-123 Cola Market 2015 Feature Tuning

Input #1
Product Features

Input #2
Feature List

Optional – not used.

Input #3
Customer Distributions

Output #1
Product Array

Output #2
WTP Matrix

#2 Simulate New Market

The Simulate Market node takes the Input Product Array and the Input Willingness To Pay (WTP) Matrix with the new Coke Zero Cherry Product. The node then predicts how many Customers will buy the new Product and the purchase decision of each Virtual Customer.

This future Market prediction (with the new Product) can then be compared to the current Market results (without the new Product). A Customer-by-Customer comparison can then be made to determine which Customers switch to Coke Zero Cherry. Further analysis then reveals which are old Coke Customers who would have bought a Coca Cola Product versus which are new Coke Customers who previously purchased Pepsi or are new to the Cola Market.

See also: CN-141 Simulate Market Node

Input #1
Product Array

Input #2
WTP Matrix

Output #1
Product Array

Analysis #1
Switched Customers

Results #1
Brand Growth

Results #2
Source of Customers

Results #3
Real Growth