Customer analysis is the essential thing for innovation and any business. It is tough to understand customer perception. Like the case with Coca-Cola:

Coca-Cola Story

Once, Coca-Cola decided to make a blind test of Coke and Pepsi. It appeared that only 40% chose Coke, so Pepsi appeared tastier. Coca-Cola decided to create a new taste that would be better than Pepsi. And Coca-Cola made it! They started to sell the new Coke.

It was the biggest mistake of Coca-Cola. People started boycotting Cola and even making protests. Once the CEO Golzueta was asked: “How can you sleep when you just sold the American Dream?”. He answered: “I sleep like a baby. I wake up every hour crying”. They Kept the old Coke.

The problem was that Coca-Cola did not know why do customers buy their product. The focus on the physical, chemical part of a product (like taste or color, or a form of a bottle) can be dangerous. Coca-cola assumed that taste was the central aspect of buying Coke, but it was irrelevant for customers.

Smart customer analysis is critical for success. We use this model to understand the demand. What makes people buy a product? People buy preferred products based on subjective perception. There is no direct 1 to 1 transmission. There are also other factors. People may perceive something, but sometimes they won’t care about this perception.

Coca-Cola assumed a product property that the problem is with the formula of Coke, and the taste must change it. However, it appeared that this perception was very weak. The taste made a little different for people. However, people wanted to drink an old taste to feel the taste of centuries.

Kano Model

is a heuristic model, a framework for the classification of product attributes. The core of the Kano Model is that the improvement of all product attributes is not a good idea.
Customers may not care about different attributes.

Washing machine company decided to make a new machine with 5000 modes. It was unnecessary because people usually use 1-3 programs. The company was proud to sell so many modes in a machine, but as you understand, no one was buying this machine, at least for a higher price than other machines.

Basic attribute

If an attribute is below the performance attribute, then the customer will not care about any improvement there; it is given for granted.

Excitement attribute

This is the essential attribute – if the company does not improve the attribute – it is not a problem; however, if the company does, customer satisfaction will skyrocket.

Dont blindly maximize the performance!

Conjoint analysis

Most used statistics used in practice. It displays the info on the importance of attributes. It allows learning if there is a linear relationship between attribute level and the resulting value for customers or there are “jumps.” E.g., it shows how many customers we lose by making the price higher.

Strengths:

1. Data collection is pretty natural – we compare the overall product
2. Not overstraining participants
3. Requirement to make a trade-off decision (“not everything is important” inflation)
4. Detailed results
5. Identification

Weakness

1. Only a small number of attributes is possible. You can use the adaptive conjoint method instead.
2. Importance of attributes depends on the span of attribute levels.
3. Assumption of independent attributes only holds for the modular products. If we make the integral approach, the soup approach, the interaction between different attributes plays a significant role. In the modular, Sushi approach, different sushi types don’t interact with each other.

Learn more in the video