Web Design User experience: Let the user really satisfied

Web page Production WEBJX article introduction: is the user really satisfied?

Since the 80 's, a variety of products and services have begun to develop around the world, replacing the early economic and market environments dominated by a large number of ordinary daily necessities. In the 90 's after the electronic and Internet industry began to develop rapidly, a variety of High-tech products with each passing day, in people's daily life everywhere. For these huge changes, we need more scientific methods to evaluate the actual benefits of industries, companies and products and services in a new economic environment, rather than just the traditional large-scale production indicators.

So, in the 90 's scientists began to target different industries in the new economic environment, more scientific and effective evaluation of different areas, different products and services, beginning to pay attention to consumer or user feedback, because whether it is the enterprise, industry, department or product, are directly facing the "customer", Customer satisfaction (satisfaction) becomes the intangible asset of these content, can directly or indirectly reflect the actual benefits, so the customer Satisfaction Index (CSI, customer Satisfaction Index) has become a new measure. Here's a quick introduction to the CSI Model:

CSI Model:
The CSI model is not a specific practical satisfaction model, but a methodology to measure satisfaction, a set of models to evaluate customer satisfaction. For a collection of general concepts that reflect customer satisfaction, it can be called "macro Model" (Macro-model), such as value, perceived quality, customer expectation, customer complaint behavior, brand/product impression, customer loyalty, etc., these concepts can be measured customer overall satisfaction of the "macro" indicators, In the model is called "Potential variable" (LV, latent Variable). The following figure is an example of a simple CSI model, with the one-way arrows giving a conceptual relationship.

(Source: C. O ' Loughlin and G. Coenders, 2002)

　
For different applications, different models are given. In 1996, for example, Woodruff and Gardial proposed the concept of value as a driving force in the relationship between product selection and satisfaction, and as a simple psychological reaction, formed a value chain. Based on such a goal, they set up a hierarchical model of customer value chain to customer satisfaction.

The literature of market research points out that for different macro concepts, more detailed factors (or elements) need to be supported, because the macroscopic concept is a vague and perceptual definition, which is usually not directly calculated. Therefore, each macroscopic concept can be decomposed into different factors, called "Microscopic Model" (Micro-model), for example, the perceived quality of products can be measured by reliability, overall evaluation after purchase, functional requirements, etc., in the model is also called "Measurement variables" (MV, Manifest Variable). A simple microscopic model can be expressed as follows:

(Source: C. O ' Loughlin and G. Coenders, 2002)

Customer Satisfaction Index model for countries (regions):

Beginning in the 90 's, the researchers began to collect data on the actual situation of countries (regions) and conduct research and establishment of satisfaction models. Among the more famous are the Swedish Customer Satisfaction Index (SCSB, Swedish customer satisfaction Barometer), the American Customer Satisfaction Index (ACSI, American satisfaction Index) and the European Customer Satisfaction Index (ECSI, European Customer Satisfaction Index). Here takes ACSI as an example to introduce.

The prerequisites for establishing the ACSI model include: 1. Satisfaction is derived from the customer's evaluation, can not be directly observed, so it is as a potential variable with multiple indicators (measuring variables) to be measured; 2. ACSI not only consider the actual consumption experience, but also pay more attention to the forecast of future prospects. The concrete macroscopic structure model is as follows:

(Source: Fornell, et al, 1996)

Because the ACSI and the macroscopic latent variables are not directly measured and need to be calculated according to several measurable variables, the following table illustrates the measurement variables in the model. (Source: Fornell, et al, 1996)

　

The establishment of the model:

ACSI model is based on the "structural Equation Modeling" (SEM, structural equation Modeling) method, to establish a different measurement variables and potential variables of the structural relationship equation, on the SEM method, you can refer to "5". In other words, the potential variables are obtained by means of measurable factors, and the overall customer satisfaction is calculated. Different models are built on the basis of different actual data. The quantified data were obtained through questionnaires. The structure and relationship of the model are also derived from the actual data, which are fitted and estimated by some mathematical algorithms (such as partial least squares pls). As the following illustration shows, before the model is determined, we assume that all the potential variables and the measurement variables have a relationship, and then use the actual data to pick out the valid variables for modeling. In addition, different measurement variables are corresponding to different questionnaires, at first we do not determine the number of effective questionnaires and content, the common method is to identify the potential variables with the most relevant questionnaire data, as a source of measurement variables. This analytical method is also known as confirmatory factor analysis (CFA, confirmatory Factor analyses), and the model is gradually determined by the constant validation model and the compatibility of the data. For more information, refer to "5".

(Reference: Ref. "2,3")

Analysis of the characteristics and effects of the model:

A scientific model of satisfaction can bring good results. Also take the ACSI model as an example. In order to measure the economic output more accurately and comprehensively in the ACSI model, and can predict economic profits, provide useful information for economic policy, become indicator of economic health, must meet some criteria: accuracy, effectiveness, reliability, predictability, coverage, simplicity, diagnosis and comparability.

These contents can be used for reference in the evaluation of product satisfaction, but need to improve the specific product characteristics and industry. Therefore, in the whole process of modeling, we must consider the potential variables/measurement variables selection, data collection, processing, algorithm application, conclusion analysis and solution extraction and so on, and make the model in the long-term landscape (between different products, Different industries) and longitudinal (different time) accumulation in the continuous optimization, making the results more comparable. Therefore, in the process of modeling, we need specific analysis and investigation for each link.

Questionnaire Design:

Here is a special mention of the questionnaire design. Questionnaire is the source of measurement variables, the data directly determines the effectiveness of the model, so in the early qualitative, the design of the questionnaire is very important. The necessary adjustments should be made under the larger structure of the model, and the design should be based on the specific circumstances of the product itself, avoiding the use of overly generic examples. The length of the questionnaire is also determined by the number of variables measured. For the measurement variables of the model, it is sufficient to select as precise and relevant a problem as possible, since even too many unrelated problems may be removed from the analysis. This requires that in the design of the questionnaire, according to the researcher's own experience and the actual situation, to filter. If the problem in the questionnaire is lengthy, it will not only improve the accuracy of the analysis result, but also may result in the increase of the customer's abandonment rate and the decrease of the reliability in the process of answering the questions.

In addition, when discussing the satisfaction index, we need to put forward several points of attention :

1. According to Oliver's definition in 1997: "Satisfaction is a reflection of the satisfaction of consumers (or users)." It is a pleasing level that a product or service provides to consumers in the sense of satisfaction associated with consumption, including below satisfaction or satisfaction. When modeling customer satisfaction metrics for a product, our target audience should be the user, not the customer. Because the user is the consumer of the product or service, and the customer is the purchaser, the customer is not necessarily the direct user of the product or service. Therefore, in the selection of target crowd, should be aware of the elimination of inappropriate "customers", so as to improve the data information and noise ratio, improve the reliability of the data.

2. Satisfaction usually has the horizontal threshold (exceeding the level of the user's psychological satisfaction) and the lower level threshold (below the level of the user's psychological satisfaction). In other words, when users get "too many good things", user satisfaction may decline, many users are focused on the horizontal threshold and ignore the upper level threshold.

3. Satisfaction is a feeling, although at the time of modeling, through various methods can make this feeling can be quantified, measurable, and make the measurement more accurate, but this is always a relatively short-term attitude, and may change with the environment and time, Therefore, the satisfaction of the product needs to be continuously based on the recommendations of the evaluation results to improve, and its model to be continuously optimized.