Pepsi Cola was introduced in 1893, 7 years after the introduction of Coca Cola in 1886.

Pepsi’s early years were full of challenges. Between 1922 and 1933 the Coca-Cola Company were asked three times to buy the Pepsi-Cola Company (they declined each time). In 1931, at the depth of the Great Depression, the Pepsi-Cola Company went into bankruptcy.

But in working their way out of bankruptcy, Pepsi-Cola’s fortune changed. In 1936, still during the Great Depression, Pepsi gained popularity through the introduction of a 12-ounce bottle. Pepsi encouraged price-sensitive consumers to switch from the 6.5-ounce bottle of Coca-Cola to the new 12-ounce bottle of Pepsi for the the same price of five cents (a nickel). From 1936 to 1938, Pepsi-Cola’s profits doubled.

The Pepsi jingle that promoted their new 12-ounce bottle ran as follows:

Pepsi-Cola hits the spot
Twelve full ounces, that’s a lot
Twice as much for a nickel, too
Pepsi-Cola is the drink for you!

See also: Wikipedia – Pepsi

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.

This Case Study uses a very common workflow pattern to build a set of part-worth Customer Distributions for each Feature.

There are two Features in this workflow:

• Cola Feature
• Bottle Feature

Each Feature has two Variations for the two Products in the Market:

• Coke Variation
• Pepsi Variation

The Feature Variations are differentiated such that Customers value the Feature offered by each Product differently. There are three types of Differentiation:

• Vertical Differentiation
• Horizontal Differentiation
• Strange Differentiation

When Features (or Products) can be rank ordered in an objective way then they are said to exhibit Vertical Differentiation. For example, there is Vertical Differentiation between a ‘1-year warranty’ and a ‘2-year warranty’ because all Customers universally agree that 2-years is better than 1-year. Similarly, Features that are distinguished by the number of gigabytes or megapixels are also examples of Vertical Differentiation as all Customers agree that the more, the better. See also: Differentiation Vertical Node

When Features (or Products) cannot be rank ordered in an objective way then they are said to exhibit Horizontal Differentiation. This means that while Customers may, on average, agree that the value of one Feature Variation is the same as the value of another Feature Variation, those Customers may disagree as to which of the two is better. For example, the value Customers place on the ‘Coca Cola’ brand might be very different to the value they place on the ‘Pepsi Cola’ brand, even though the taste of the two Products is hard to distinguish. See also: Differentiation Horizontal Node

Strange Differentiation relates to the Standard Deviation (SD) among the Customers Willingness To Pay (WTP). An example of Strange Differentiation is when a Product lacks the appeal of the mass market, but is nonetheless valued by a select niche of Customers. As Mean and SD are always paired together, the degree of Strange Differentiation is also set within the ‘Differentiation Vertical’ node. See also: Strange Differentiation

Features can have four types of Variations:

1. Quality = adjusts the average amount of Vertical Differentiation relative to the norm (shifts the Mean of the part-worth Customer Distribution)
2. Niche = adjusts the amount of Strange Differentiation (shifts the Standard Deviation)
3. Expense = adjusts the relative Cost of the Feature
4. Conformity = adjusts the amount of Horizontal Differentiation relative to the norm (shifts the Correlation between the part-worth Customer Distributions)

In this Case Study of the early Cola Market, Coca Cola enjoys significant Vertical Differentiation as Customers set the value of the Pepsi Cola Variation at almost half the value of Coca Cola (Quality = -1.0 would set the average WTP to be 50%, but here Quality = -0.8). Pepsi also doesn’t have the economies-of-scale that Coke enjoys, so it’s Cost is 25% higher (Expense = +0.5). But Pepsi has some Strange Differentiation (Niche = +0.2), and the Pepsi Flavor provides some Horizontal Differentiation (if Conformity = +1.0 then Customers would consider the Coke and Pepsi Flavors to be identical, but here Conformity = +0.9).

The Mean value of the part-worth Customer Distributions is defined in the ‘Differentiation Vertical’ node (Coke’s Cola Mean = 5.0). The Standard Deviation is also defined in the same node (Coke’s Cola SD = 0.5 x Mean = 2.5), as well as the Feature’s Cost (Coke’s Cola Cost = 0.25 x Mean = 1.25).

The Correlation Matrix between the part-worth Customer Distributions for all of the Feature Variations is defined by the ‘Differentiation Horizontal’ node.

When the Vertical Differentiation and the Horizontal Differentiation for the Feature has been defined, a set of part-worth Customer Distributions can be created with the Matrix Distributions node. The part-worth Customer Distributions contain the Willingness To Pay (WTP) that each Customer has for each Feature Variation. See also: Matrix Distributions Node

Click on an icon to see and scroll through the enlarged version of the images.

The Vertical Differentiation and the Horizontal Differentiation of Pepsi’s Bottle is defined using the same Differentiation Workflow Pattern as above.

In this case, the value of the cola in Pepsi’s larger bottle is higher for Customers (but not twice as high), and the increased Strange Differentiation (Niche) is representative of the additional appeal Pepsi has to some Customer Demographics. The extra Expense of the larger bottle will also drive up Pepsi’s Product Cost.

The two Cola Products in the Market are generated from the part-worth Customer Distributions of the Features:

• Coke (Price = 5 cents) = Cola.Coke + Bottle.Coke
• Pepsi (Price = 5 cents) = Cola.Pepsi + Bottle.Pepsi

Coke has the greater Mean, indicating that it has Vertical Differentiation.

Pepsi has the greater Standard Deviation (SD), indicating that it has Strange Differentiation.

The Correlation between the Willingness To Pay (WTP) Distributions for the two Products is about 0.90, indicating that both enjoy some Horizontal Differentiation from the other.

The Profit Engine node, at the end of the workflow, predicts that Pepsi’s Market Share would rise to about 37% in this market, with Coke taking the remaining 63%.

These results are roughly inline with reality as the Market Share for Coca-Cola after World War II was about 60%.