Customer analysis is the¬†essential thing for¬†innovation and¬†any business. It is tough to¬†understand customer perception. Like the¬†case with Coca-Cola:
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: ‚ÄúHow can you sleep when you just sold the¬†American Dream?‚ÄĚ. He answered: ‚ÄúI sleep like a¬†baby. I wake up every hour crying‚ÄĚ. 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‚Äôt 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.
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.
If an¬†attribute is below the¬†performance attribute, then the¬†customer will not¬†care about any improvement there; it is given for¬†granted.
This is the¬†essential attribute¬†‚Äď if the¬†company does not¬†improve the¬†attribute¬†‚Äď it is not¬†a¬†problem; however, if the¬†company does, customer satisfaction will skyrocket.
Dont blindly maximize the¬†performance!
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 ‚Äújumps.‚ÄĚ E.g., it shows how many customers we lose by¬†making the¬†price higher.
- Data collection is pretty natural¬†‚Äď we compare the¬†overall product
- Not¬†overstraining participants
- Requirement to¬†make a¬†trade-off decision (‚Äúnot¬†everything is important‚ÄĚ inflation)
- Detailed results
- Only a¬†small number of¬†attributes is possible. You can use the¬†adaptive conjoint method instead.
- Importance of¬†attributes depends on¬†the¬†span of¬†attribute levels.
- 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‚Äôt interact with each other.
Learn more in¬†the¬†video