Factor Analysis is a technique used in marketing. It helps solving problems where a lot of information can be grouped together. For instance measuring quality as a whole or in more detail like taste, design and customer service. The analysis can be based on actual data or people opinions.
Factor Analysis in Marketing using a basic example.
1. The Problem
Let’s take quality for example. Not all your customers use your product in the same way, so each of them may have a different understanding of its quality. If you want to improve it then you need that (understanding). So, you need to figure out what they mean by quality so that:
- Makes sense to you
- It is in a form that you can use it for decisions (improving, investing etc)
That can be made easier with Factor Analysis.
Let’s say you want to understand what your customers mean by quality. After having a discussion you get many opinions on, say, customer service.
Others talk about how nice the operator talked to them, others talk about how understanding the person on the phone was with their problem while others describe the operator as kind when dealing with their problem.
All these aspects refer to empathy. However, not all customers are familiar with this term, neither can they all spot it when they experience one.
So, when the time comes to measure empathy of customer service operators as part of customer service quality, you need to capture many possible views. However, asking your customers about all these different things, when in the end you want to measure empathy, may push them away and put the project in risk of low response rate.
2. The Solution
So, you can test it to a few people, then use Factor Analysis to decide on which questions to keep. In this case Factor Analysis is used for reducing questions in your survey.
Another interesting use of Factor Analysis is to discover new factors. Imagine that from your discussion with your customers, you got new information that could not fit with what you knew already about your product.
For example, they are talking about its size, its packaging, its ability to fold etc and all they refer to is the fact that it saves them space. In the case where you never thought about that you may look for the performance of “space-saving” feature in your product and that of the competition.
If you combine this with the Customer Value Analysis, then you can look for benefits in differentiating there. What do your customers value at your product that you never thought of?
3. The Process
How can I actually perform Factor Analysis?
There are 2 popular options shown here.
The first one is advised to be used as a pre-test.
The second one is a more complex technique that creates the final questions to be used based on the answers and as a consequence the total measures to analyze are merged into “factors”.
You can use the correlation index to check whether two questions may be actually referring to the same thing. This way you can select which one to keep. An easy to use analytical tool is the correlation matrix.
2. Principal Components Analysis
This technique is used for reducing the number of answers by combining these according to their effect on the. For example, a customer’s score on customer service operator’s kindness, nice behaviour and understanding of their problem are probably found to be better grouped. The grouped score based on business sense can represent empathy. The empathy score can be then used for further analysis.
The free Real-Statistics tool for MS-Excel can handle such analysis. This toolbox supports the Pearson correlation index to create the correlation matrix as well as Spearman and Kendall’s Tau. You can find the process for the particular software here. A guide on Principal Components Analysis using the same tool can be found here