# How to calculate customer satisfaction level

Measuring customer satisfaction using customer loyalty software is an important way for an institution or company to determine the success of its efforts to satisfy customers. Usually, customer satisfaction is assessed not only as one attribute but as multiple attributes for companies engaged in services or products. Even government agencies that deal directly with public services periodically calculate the level of satisfaction of the people served.

How to calculate the level of customer satisfaction can be started by making a questionnaire consisting of the level of importance and level of satisfaction. Consumers are asked to rate between 1 and 5 the importance and performance level of each predetermined attribute. For example, at a basic training event, participants are asked to rate the performance of the organizers. The values or attributes used include cleanliness of the class, friendliness of the staff during registration, diversity of food menus, etc. All attributes are asked two things: how important the attribute is and how well it performs. All were measured from 1 (very unimportant or very poor) to 5 (very important or very good).

Both dimensions can be processed using quadrant analysis. This quadrant analysis (importance performance analysis) actually combines two dimensions with a diagram so that it becomes input for organizers on which attributes must be improved. I have explained this quadrant analysis in an article entitled: Evaluation Techniques for Mentoring or Counseling Activities.

The quadrant analysis does not yet reflect the combination of the two dimensions (importance and performance); the explanation of the quadrant analysis is still separate, so it does not answer whether the training participants are satisfied with the overall service provided by the committee or not. The method of combining and formulating conclusions from the two dimensions will be discussed using the two methods below.

## Two methods of measuring customer satisfaction: customer loyalty software

The methods in question are the customer satisfaction index, or CSI, and the Fishbein method. The difference?
CSI is the level of customer satisfaction based on certain attributes, while fishbein expresses an assessment of a particular object based on beliefs summarized from the corresponding attributes. In essence, CSI expresses the attitude of how satisfied the customer is with the object, while Fishbein already states the attitude, that is, how much the customer likes the object. It can be interpreted that Fishbein is deeper in its analysis because it has stated the attitude of liking or not and because there is an interaction between performance and importance assessments.

## Customer Satisfaction Index (CSI)

The customer satisfaction index method (Customer Satisfaction Index) is an index to measure the level of customer satisfaction based on certain attributes. According to Dixon (1991), there are four steps in calculating the Customer Satisfaction Index (CSI), namely:

### Determine the mean importance score (MIS) and the mean satisfaction score (MSS).

This value is scaled from the average level of importance and performance.

Where:
n = number of respondents
Yi = The importance value of the i-th attribute
i = Performance value of ith attribute

### Creating Weight Factors (WF)

This weight is the percentage of MIS value per attribute to the total MIS of all attributes.

### Creating Weight Score (WS)

This weight is a multiplication of the Weight Factor (WF) with the average level of satisfaction (Mean Satisfaction = MSS)

WSi = WFi x MSS
Where: i = ith attribute

Calculating Weighted Total (WT), which is the sum of the weighted scores of all variables.

Determining the Customer Satisfaction Index (CSI), dividing the weighted total by the nominal scale used and then multiplying by 100 percent. The CSI formula is as follows:

Where:
P = number of importance attributes
5 = number of scales

Customer Satisfaction Index (CSI) Value Criteria

## Fishbein’s Multiattribute Attitude Model

Fishbein’s attitude model suggests that the assessment of a particular object is based on a set of summarized beliefs about the attributes of the object in question which are weighted by the evaluation of the attributes. This model states that a consumer’s attitude towards an object will be determined by his attitude towards the various attributes possessed by the object. The ei component measures the evaluation of the importance of the attributes possessed by an object. Meanwhile, bi measures consumer confidence in the attributes possessed by an object.

Symbolically the Fishbein multiattribute attitude model formula is formulated in the formula:

Where:
Ao = consumer’s overall attitude towards the object
bi = the strength of consumer beliefs that have attribute i
ei = consumer evaluation of attribute i
n = relevant attributes (dimensions)

After identifying the dimensions, the appropriate bi and ei measurements are made. The ei component describes attribute evaluation, which is a component that explains how much the consumer’s attitude towards the attribute as a whole Evaluation is typically measured on a 5-point evaluation scale that ranges from very important, important, moderately important, unimportant, and very unimportant.

For example:
Expectation Level:

This evaluation will be carried out for each of the attributes studied. The bi component describes how strongly consumers believe that a modern or foreign restaurant has a given attribute. The attributes used for the bi component should be the same as the attributes used to calculate the ei component. Belief is measured on a 5-point scale of realized possibilities that range from very good to not good. For example:

Service performance in responding to complaints:

Average responses are then calculated for bi and ei ranging from a maximum score of 5 to a minimum score of 1. To estimate the attitude assessment of the restaurant using the åbiei index, each belief score (bi) must first be multiplied by the corresponding evaluation score (ei). Then all multiplication results must be summed, resulting in a total consumer attitude assessment score. The assessment of consumer attitudes towards restaurants can be compared with the maximum total score of the existing evaluation components, namely by multiplying the ideal trust score (bi) by the existing evaluation score (ei).

Before we give an interpretation of the results of the consumer assessment, first determine the range of the rating scale. Also determine the minimum score and maximum score of the assessment that may be given by consumers.

Scale range formula:

Where:
m = the highest number in the measurement
n = lowest number in the measurement
b = number of interpretation classes available

The range for evaluation (importance) and confidence level (implementation) is:

After the interval is known, a scale range is determined based on the level of importance:

1.00-1.80=Very important
1.81-2.60=Not important
2.61-3.40= Ordinary
3,41-4,20 =Important
4,21-5,00= Very important

While the division of classes based on the level of confidence is:

1.00-1.80 = Very unfavorable
1.81-2.60 = Unfavorable
2.61-3.40 = Ordinary
3.41-4.20 = Good
4.21-5.00 = Very good

After knowing the importance and trust, then the attitude value (Ao) is obtained which is the multiplication of the level of importance and the level of trust (Simamora, 2004). The magnitude of the range for the attitude category is:

so the division of classes based on the attitude value (Ao) is:

1.00-5.80 = Strongly dislike
5.81-10.60 = Dislike
10.61-15.40 = Ordinary
15.41-20.20 = Like
20.21-25.00 = Very like

While the total attitude value (Ao), the range value is obtained from:

(The number 15 is obtained from the number of attributes or the number of variables asked by respondents. In the video example below, the number of attributes is 16, so the multiplier is also 16)

so that the total attitude class division (Ao) is obtained, as shown in the following table:

Total Attitude Value Range (A0)

The range of values starts from 15 as the minimum number because in the first respondent, there are 15 attributes collected.  So that for the first respondent, the minimum value if he answers 1 on all attributes will be 15.

while the maximum number is obtained from 25 x 15 = 375. the number 25 is obtained from the maximum number that can be obtained from multiplying the level of importance (maximum value 5) with the level of performance (maximum value 5). the number 15 is obtained from the number of attributes or variables assessed. in the video example below the number of attributes is 16.

Basically, Fishbein is about adjusting the scale that adjusts the number of attributes. So that the new value of the multiplier of the level of importance and the level of performance produces one value which of course has a different scale value to determine the attitude.

For more details, watch the video after this article, the fishbein calculation in this video continues the previous CSI video. (the number of attributes or variables in the video is 16):