Rethinking Satisfaction Research from Now on

User satisfaction is a compulsory introductory course for us to conduct market research and user research. After evolution, user satisfaction is also constantly changing, and we need to re-examine the satisfaction research.

Satisfaction research can be said to be a required course for every market research/user researcher, and it is the most "basic" and common project of all project types. But in fact, satisfaction research is not simple at all. On the contrary, it has a history of nearly 60 years and has gone through many iterations.

01 Major Iterations of Satisfaction Research

First Generation: Service Standards and Employee Code of Conduct

Satisfaction research emerged in the 1970s. The earliest literature can be traced back to "An Experimental Study of Customer Satisfaction, Expectations, and Satisfaction" published by American scholar Cardozo in 1965. It was he who first introduced the concept of "customer satisfaction" into the business field, and made companies realize the importance of service quality and begin to accept and apply market research on service quality.

The satisfaction study at this stage focuses on the standardization of the service process, and ensures that employees follow the established service standards through surveys. This type of survey is called a "service implementation survey", and its purpose is to improve employees' service behavior and ensure the implementation of service standards through inspection and feedback mechanisms.

Service implementation surveys are usually conducted using questionnaires or mystery shopping. The former is lower cost but has weaker evidentiary power, while the latter obtains more powerful assessment evidence by simulating real service scenarios.

Second Generation: Perceived Quality and Service Effectiveness Evaluation

By the mid-1980s, scholars began to realize that there was a difference between consumers’ perception of service quality and companies’ understanding. At this time, satisfaction research turned to an evaluation system centered on customer perception, namely “perceived quality survey”.

This survey not only focuses on the service process, but also pays more attention to the customer's direct feelings about the service effect. It measures the overall effect of the service by evaluating the perceived quality of the service, covering both the front-end direct service and the back-end support department, forming a comprehensive service evaluation system.

In addition, perceived quality surveys use statistical techniques to identify key influencing factors, helping companies optimize resource allocation and improve service shortcomings.

The third generation: Satisfaction index model and macro service evaluation

In 1988, the new satisfaction model proposed by American scholar Fornell based on structural equations pushed satisfaction research to a new height and became the basis for countries around the world to formulate national satisfaction index models.

Professor Fornell and his team first built a causal SCSB evaluation model for Sweden, and then developed the customer satisfaction model ACSI, the European customer satisfaction model ECSI, and the Chinese customer satisfaction model CCSI on this basis.

The Swedish Customer Satisfaction Model (SCSB), established in 1989, is the earliest national customer satisfaction index model. The model consists of five key elements, among which customer satisfaction is the core, customer expectations and perceived value are influencing factors, and customer complaints and loyalty are the manifestation of results.

It is worth noting that the evaluation of customer satisfaction in the model is not based on a single purchase experience, but on the customer's long-term accumulated purchase experience and overall evaluation. The structure of the model is as follows:

The American Customer Satisfaction Index (ACSI) was established in 1994 after being improved on the SCSB model. It was based on the actual needs of enterprises and consumers, introduced the concept of perceived quality, and added consideration of customer expectations, constructing a more comprehensive and practical evaluation system.

The model consists of six main structural variables, as shown in the figure, and it has been widely used and deeply studied in academia. The structure of the ACSI model is as follows:

The European Customer Satisfaction Index (ECSI) was constructed in 2000 on the basis of ACSI and organized by the European Quality Organization, the European Quality Management Foundation and other institutions.

It has made innovations in customer satisfaction research by removing the variable of customer complaints and introducing corporate image as a key factor affecting customer satisfaction, making the model more comprehensive. The model contains six main variables, and the relationship between them is shown in the figure below:

The establishment of the China Customer Satisfaction Index (CCSI) started relatively late. In 1997, under the promotion of the China Quality Association and the National User Committee, the CCSI system research began, and joint efforts were made with top domestic academic institutions such as Peking University, Renmin University, Tsinghua University, and the Chinese Academy of Social Sciences. On the basis of referring to and learning from the ACSI, the design work of a national satisfaction index model suitable for China's national conditions was carried out.

Compared with the ACSI model, the CCSI model adds brand information and removes customer complaint variables. Its characteristic is that it calculates a percentage score to show user satisfaction.

ACSI is still the most complete and best-applied national customer satisfaction theoretical model.

However, the ACSI model was originally created to measure the quality of economic output from a macro perspective and monitor the operation of the macroeconomy. It also considers customer satisfaction comparisons at the national and industry levels, rather than providing diagnostic guidance for specific companies. Therefore, companies still need to combine the perceived quality model when applying it to ensure the comprehensiveness and depth of the questionnaire.

02 Development of Satisfaction Research in China

China's commercial satisfaction research practice can be traced back to the mid-1980s to the late 1980s, when foreign companies represented by Procter & Gamble and market research companies represented by AC Nielsen were the first to introduce customer satisfaction research.

However, early customer satisfaction research mainly focused on service implementation and service effectiveness perception, that is, at the micro level of the enterprise. It was not until 1997 that the country began to promote the research and development of the China Customer Satisfaction Index (CCSI), and around 2000 China began to carry out large-scale macro-industry customer satisfaction research.

The telecommunications industry was one of the earliest industries to practice the structural equation satisfaction model and incorporate satisfaction measurement scores into KPIs. The telecommunications industry also has its own index called the Telecommunications Customer Satisfaction Index (TCSI). However, no matter how it changes, the underlying model is based on the ACSI model.

1. Research on service implementation

After entering the Chinese market, foreign-funded enterprises such as fast-moving consumer goods and automobiles, represented by Procter & Gamble, mainly conduct market research projects such as channel mystery shopper inspections aimed at checking service implementation in order to build a sales channel system covering the whole country.

The following is a case in the automotive industry, which comprehensively tests the sales service system of a certain automobile 4S store from both hardware and software aspects. It evaluates whether a certain brand of automobile dealership has implemented the prescribed service requirements of the sales process in the sales operation process, and at the same time finds out the areas that need reference and improvement in the management of the brand of automobile dealership.

This type of project has very high requirements for offline execution resources and experience. The cost of a single project is very high. Basically, only the leading brands in each category have the money to do it, and once it is done, it needs to be done continuously.

To this day, there is still demand and market for this type of service implementation inspection and research, and it is still a necessary investment for many companies with window departments (4S stores, chain stores, counters, etc.).

2. Perceived Quality Survey

Collect subjective evaluations of each touchpoint from the customer's perspective.

From 2000 to 2013, CATI telephone interviews (questionnaires) were popular in China. Their characteristics were short questionnaires of less than 15 minutes and large sample sizes, often involving tens of thousands of samples. After 2013, online questionnaire surveys gradually became the mainstream.

The perceived quality survey mainly uses the hierarchical analysis method model. Simply put, the first-level indicators are divided into second-level indicators, and the second-level indicators are further divided into third-level indicators.

Examples:

This approach is now commonly used within many companies, including leading manufacturers, to obtain user satisfaction with key variables and factors through direct questioning, even if it is a latent variable and unobservable, such as the secondary indicators in the above case.

There are benefits to doing this; it is simple, flexible and highly operational. However, the problem is that it can only solve the problem of improving the experience at the micro-product level and cannot be linked to business operations or financial indicators. In other words, this method cannot conduct macro-level research on the company and can only solve the problem of improving the touchpoint experience, not the strategic level.

3. Satisfaction Index Model

Since 2000, corporate clients in many industries such as fast-moving consumer goods, automobiles, and telecommunications have generally hired third-party companies to conduct satisfaction surveys based on structural equation models. They are eager to determine the company's priority improvements from a strategic level, and have also put forward requirements for the optimization of every unit of the company's market competition factors.

For a time, every company had its own satisfaction model, but the models of different companies varied greatly, and there was a mixture of good and bad.

At the same time, as Party A companies build their own user research teams, some researchers also try to use structural equation models to build models/indicator systems. However, most of these models/indicator systems have not been tested in theory and practice, which often results in unstable models and unexplainable data.

03 We need to re-examine satisfaction research

What enterprises really need today is a solution that combines satisfaction research at the enterprise's macro-strategic level with a composite indicator system for improving product experience at the micro level.

However, satisfaction research at the macro-strategic level needs to be supported by models that have been verified in theory and practice and have common industry characteristics. It cannot be done by simply running a structural equation model. Otherwise, the country would not have organized top domestic academic institutions to jointly tackle the problem. The ACSI model is the most mature and effective one, but there are not many people in China who have truly mastered the ACSI model technology patents.

The next step is how to link the satisfaction research at the macro-strategic level with the product satisfaction research at the micro-level, which also requires professional skills and experience.