Customer Satisfaction - Science topic

Dear Bill,

I hope you will find the following article and the enclosed references are useful for you research.

Best regards

Hamdi

***********

Disparities in Customer Satisfaction and Loyalty Metrics Post-COVID-19: Analyzing the Recovery of Service CSAT and the Lagging NPS":

The recovery of Service Customer Satisfaction (CSAT) to pre-COVID levels, contrasted with the lagging Net Promoter Score (NPS), presents a complex scenario that warrants exploration into the underlying factors. Research indicates that while CSAT reflects immediate customer experiences and satisfaction with services, NPS captures broader loyalty and willingness to recommend, which may be influenced by different dynamics in the post-pandemic landscape.

One potential reason for the divergence in recovery rates between CSAT and NPS could be the evolving expectations of customers in the wake of the pandemic. As noted in studies, healthcare professionals have had to adapt their services to meet the diverse expectations of patients recovering from COVID-19, which has led to generally positive reviews of post-COVID services (Sunkersing, 2024). This adaptability may enhance immediate satisfaction (CSAT) but does not necessarily translate into long-term loyalty or advocacy (NPS). Furthermore, the pandemic has introduced significant stressors and uncertainties, which can affect customer perceptions of service reliability and trustworthiness, thereby influencing NPS more adversely than CSAT (Shaikh et al., 2023; Menges et al., 2021).

Moreover, the impact of service recovery strategies during the pandemic has been highlighted as a critical factor. Research suggests that effective service recovery, including timely communication and compensation, can mitigate the negative effects of service failures experienced during the pandemic (Mazhar et al., 2022). However, the effectiveness of these strategies may vary across different sectors, influencing CSAT and NPS differently. For instance, in the hospitality sector, customer satisfaction has been closely tied to the perceived quality of service recovery efforts, while long-term loyalty may hinge on broader brand perceptions and experiences ("Influence of Domestic Tourism Marketing Strategies on Post Covid-19 Hospitality Sector Recovery in Isiolo County", 2024).

Additionally, the psychological and emotional toll of the pandemic on consumers cannot be overlooked. Studies have shown that many individuals experience lingering effects post-COVID, such as anxiety and fatigue, which can influence their overall satisfaction and loyalty to services (Shaikh et al., 2023; Rao et al., 2021). This suggests that while immediate service interactions may be satisfactory, the overall emotional and psychological context may hinder the establishment of strong loyalty metrics like NPS.

In light of these findings, it would be prudent for organizations to commission further research to delve deeper into the nuances of customer satisfaction and loyalty in the post-COVID era. Understanding the specific factors that contribute to the disparity between CSAT and NPS could inform strategies to enhance customer loyalty and improve long-term business outcomes. This could include exploring the role of service quality, recovery strategies, and the psychological impacts of the pandemic on consumer behavior (Zygiaris et al., 2022; Chan et al., 2023).

In conclusion, while CSAT has rebounded to pre-COVID levels, the lagging NPS suggests a need for a more nuanced understanding of customer loyalty in the post-pandemic context. Future research should aim to unpack these complexities and provide actionable insights for businesses navigating this new landscape.

References:

1. (2024). Influence of domestic tourism marketing strategies on post covid-19 hospitality sector recovery in isiolo county. Journal of Hospitality and Tourism Management, 7(1), 106-118. https://doi.org/10.53819/81018102t5313

2. Chan, K. H., Yen, Y. Y., & Ng, P. C. (2023). What motivated malaysians in banking with the service provider in the post covid-19 pandemic era?. Issues and Perspectives in Business and Social Sciences, 3(2), 38-47. https://doi.org/10.33093/ipbss.2023.3.2.2

3. Mazhar, M., Ting, D. H., Hussain, A., Nadeem, M., Ali, M. A., & Tariq, U. (2022). The role of service recovery in post-purchase consumer behavior during covid-19: a malaysian perspective. Frontiers in Psychology, 12. https://doi.org/10.3389/fpsyg.2021.786603

4. Menges, D., Ballouz, T., Anagnostopoulos, A., Aschmann, H. E., Domenghino, A., Fehr, J., … & Puhan, M. A. (2021). Burden of post-covid-19 syndrome and implications for healthcare service planning: a population-based cohort study. Plos One, 16(7), e0254523. https://doi.org/10.1371/journal.pone.0254523

5. Rao, S., Amara, V., Chaudhuri, S., Rao, B. K., & Todur, P. (2021). “post-covid-19 syndrome:” the new pandemic affecting healthcare workers and how the frontline warriors are battling it. Indian Journal of Palliative Care, 27, 313-318. https://doi.org/10.25259/ijpc_160_21

6. Shaikh, S., Siddiqi, Z., Ukachukwu, C., Mehkari, Z., Khan, S., Pamurthy, K., … & Brown, A. (2023). Covid-19: post-recovery manifestations. Cureus. https://doi.org/10.7759/cureus.36886

7. Sunkersing, D., Ramasawmy, M., Alwan, N. A., Clutterbuck, D., Mu, Y., Horstmanshof, K., … & Heightman, M. (2024). What is current care for people with long covid in england? a qualitative interview study. BMJ Open, 14(5), e080967. https://doi.org/10.1136/bmjopen-2023-080967

8. Zygiaris, S., Hameed, Z., Alsubaie, M., & Rehman, S. (2022). Service quality and customer satisfaction in the post pandemic world: a study of saudi auto care industry. Frontiers in Psychology, 13. https://doi.org/10.3389/fpsyg.2022.842141

You can applied the Kano Model to positioning your products/services, then differrentiation with it attribute; Basic, Performance and Delight.

After that evaluation the influence of each attribute to Diamond jewelry customer satisfaction by SEM.

Dear Ayoub Kechiche,

If you want to set up a questionnaire for your investigation, you must know your conceptual model with the relationships of variables. As per your title, there is one independent variable (CRM), and its impacts on the two dependent variables (customer satisfaction and loyalty). If I am wrong, please suggest me about it.

Further, you must understand the dimensions and indicators of the variables in your study to set up your questionnaire. If these things are clear, I can help you devising your questionnaire.

Expecting your positive response if you need help from me. Thank you.

yes sir, I have messaged the author, on IEEE the full paper access is not there. Thanks

Yes. It influences customer satisfaction. As an example in restaurants, based on the hotly served delicious dishes to the neighbors, a particular food item orders increased for a steward by other customers. In a showroom, because of the central AC set up. reduces customer annoyance to wait to resolve a complaint.

Researching the integration of artificial intelligence (AI) in financial services in Uganda presents an opportunity to understand the impact of technological advancements on customer satisfaction and trust in a rapidly evolving digital landscape. Here's a structured approach to exploring this research question:

By following this structured approach, your research can provide valuable insights into the effects of AI integration on customer satisfaction and trust in Uganda's financial services sector, contributing to both academic knowledge and practical applications for industry stakeholders.

A relatively universal metric of project success can be measured by 3 core project success factors: 1) Cost (was it within budget) 2) Schedule (was it completed on time) and 3) Performance (did the project accomplish its objectives). I was taught that in my project management classes, and evaluated as such upon the completion of my (and my fellow PM colleagues) projects (ranging in scope from several thousand- to many-millions of dollars) during my corporate career.

Yes, I ordered a device from them - and paid for it. The device was never shipped, and after a while they stopped answering my emails.

Generative AI can enhance the customer experience in e-commerce in many ways, including personalised recommendations and content, interactive preview/try-on, chatbots and virtual assistants, fraud detection, and customer feedback analysis. Some questions for a survey for online customers could be "How often do you use e-commerce services?", "Which of the following e-commerce applications of generative AI have you encountered / would like to use (choose all that apply)?" and "If you have used applications of generative AI in e-commerce, how did they work for you? How could you improve these applications?"

"Artificial intelligence (AI) is no longer confined to mere automation; it is now an active participant in the pursuit of knowledge and understanding. However, in the realm of scientific research, the integration of AI marks a significant paradigm shift, ushering in an era where machines and human actively collaborate to formulate research hypotheses and questions. While AI systems have traditionally served as powerful tools for data analysis, their evolution now allows them to go beyond analysis and generate hypotheses, prompting researchers to explore uncharted domains of research..."

intellectual capital is a valuable asset for organizations striving to implement sustainable practices. It enables innovation, knowledge sharing, stakeholder engagement, and effective management of sustainability-related challenges. Organizations that invest in and leverage their intellectual capital are better positioned to achieve their sustainability goals and reduce their environmental impact.

Actually you are asking about the scale of the factors which impact perceived service quality, then the perceived service quality will impact the customer satisfaction together with some other factors. However, in some research studies, pp tried to argue that factors which impact perceived service quality can impact directly customer satisfaction. So the answer should be the "scale" for "perceived service quality" and for "customer satisfaction" are quite different

The same question I asked ChatGPT, here is its answer:

Blending two theoretical foundations can be a complex task, but here are some steps that may help you to blend Planned Behavior Theory and Relationship Marketing Theory:

Identify the key concepts of each theory: Start by understanding the key concepts of each theory, including the main constructs, assumptions, and propositions.

Look for areas of overlap: Look for areas of overlap between the two theories, such as shared concepts or similar assumptions. For example, both theories recognize the importance of attitudes and beliefs in predicting behavior.

Identify areas of conflict: Identify areas where the two theories conflict or have differing assumptions. For example, Relationship Marketing Theory may emphasize the importance of long-term relationships, while Planned Behavior Theory focuses on individual decision-making.

Explore ways to reconcile conflicting ideas: Look for ways to reconcile conflicting ideas between the two theories. For example, you might consider how Relationship Marketing Theory's emphasis on relationships could influence an individual's attitudes and beliefs, and how this could then impact their behavior.

Develop a new model: Based on your analysis, develop a new model that integrates the key concepts of both theories. This could involve modifying one or both of the original theories to create a new model that incorporates the most relevant and useful concepts.

Test the new model: Finally, test the new model through empirical research to determine its effectiveness in explaining or predicting behavior. This could involve collecting data from individuals or organizations and analyzing the results using statistical methods.

Overall, blending two theoretical foundations requires a careful analysis of the key concepts and assumptions of each theory, and an understanding of how they can be integrated to create a new, more comprehensive model.

Here is an example questionnaire on internal customer satisfaction using "Internal Customer Satisfaction" as the dependent variable:

1 - Not Satisfied 2 - Slightly Satisfied 3 - Neutral 4 - Satisfied 5 - Very Satisfied

1 - Not Responsive at All 2 - Somewhat Responsive 3 - Neutral 4 - Responsive 5 - Extremely Responsive

1 - Poor Quality 2 - Below Average Quality 3 - Average Quality 4 - Above Average Quality 5 - Excellent Quality

1 - Very Unprofessional 2 - Somewhat Unprofessional 3 - Neutral 4 - Somewhat Professional 5 - Very Professional

1 - Not Reliable at All 2 - Somewhat Reliable 3 - Neutral 4 - Reliable 5 - Extremely Reliable

1 - Not Satisfied 2 - Slightly Satisfied 3 - Neutral 4 - Satisfied 5 - Very Satisfied

1 - Not Likely at All 2 - Somewhat Unlikely 3 - Neutral 4 - Somewhat Likely 5 - Very Likely

Thank you for participating in this survey! Your feedback is important to us.

This is just an example of kind of questions you can create in your questionnaire you can easily do it using Google Forms most easy way

Hope this helps

Zinthel and parsuramas 5 point locker scales techniques can be used for satisfaction

Here are some potential research interview questions on customer satisfaction, along with reasons for why these questions might be useful:

Dear Meron Belayneh,

I am a section editor at the

International Journal of Management & Entrepreneurship Research . You can send your paper by consulting the journal page or send it to this email address: editor@fepbl.com

The application of artificial intelligence (AI) in hospitals yields many advantages but also confronts healthcare with ethical questions and challenges.

Artificial intelligence in hospitals: providing a status quo of ethical considerations in academia to guide future research - PubMed (nih.gov)

(PDF) ARTIFICIAL INTELLIGENCE IN HOSPITALS (researchgate.net)

I am interested

The following RG link is also very useful:

Mohammad Kamrul Ahsan in general the number of respondents needs to be between 300 and over. For your side, 600 and over (600 - 800 respondents) may be convenient for the research. 365 respondents had been used in my research.

I am done answering your survey.

Piyush Sharma thank you a lot!!! That is super helpful!

I need it too, Please if you get let me know

Check out the information available from the American Society for Quality.

You may like to join the organization as well for more access to information and networking with other quality professionals related to that specific application.

Hopefully the following publications could help:

Try to target respondents more exactly...

The following research directions and the related publications may further help:

Hello Martin,

The simple answer is, model 1 includes just the so-called control variables as IVs (independent variables). Model 2 adds the target IVs to that set.

The extent to which model 2 (and therefore the target IVs) substantively add to the explanatory power of model 1 (just the control variables) may be determined by the difference in R-squared (proportion of variance in the DV that is explained/accounted for by differences in the IVs for a model) between model 2 and model 1. A "big" difference would suggest that the target IVs help to explain differences in the DV above and beyond what the control variables are able to explain/account for.

This approach is sometimes called hierarchical regression.

Good luck with your work.

Hii Furong Jia, I think it is vast a question and it depends on what you are researching. But in general, customers now want hyper-personalised experiences more than ever, and big data analytics provides the findings to make this possible. One example would be NETFLIX, the company focuses hugely on market reality ( providing only what viewers want before the viewers ask for it. Therefore, enhances the whole experience and leads to high satisfaction.

I am also interested to the answers of this topic.

Md. Qaiser Reza Dear Mr Qaiser Reza;

Thank you for your response. That link is a very useful source, and I myself had participated in the answers too. But I am to find the acceptable error range exclusively in the human behavior field.

Here you need something called weighting, try to check any reference is statistical methods or design of experiments to understand the steps for the equation.

Your satisfaction survey must also consider attributes of the evaluated service described by the target audience of your survey. Otherwise, if your variables were elaborated based on a third-party scale, you will have a much larger error than the statistic.

Dear Valeria Demchenko,

This type of relationship can be studied by analyzing the correlation between the change in the level of spending on advertising campaigns presenting the company's product and / or service offer, the level of effectiveness and efficiency of these campaigns, and the change in the level of financial revenues from sales of the company's offer. In addition, it is also important to take into account periodically conducted market research in order to diagnose changes in consumer preferences, including changes in citizens' awareness of the brand and mission of a specific company, assessment of its activities and product and / or service offer.

Best wishes,

Dariusz Prokopowicz

In the last year there's been a surge in e-sports consumption, including some major players as newcomers (F1). You could look at how the COVID pandemic has changed the way people consume games (including Twitch), meaning we no longer just "play" but also share them and see competitions.

Hermann Lassleben Based on the comments, your work might have some latent variable.

Good luck.

Thank you all!!!!! I have added it as a moderator finally for what I aim to test.

Regards

Kara

Thanks a lot, Oliva M. D. Martins, Karen Bayne, Zubair Nawaz, Gonçalo Rodrigues Brás for your valuable contributions.

George Markides Total Revenue, Profitability Rate, and Cost Percentage

I agree with @ Gautam Mahajan Sir

Quality management improves final customer satisfaction

Passenger satisfaction is not guaranteed if the quality of the services that public transit delivers do not result in increased passenger perceptions of value in relation to the fares paid. Therefore, providing passenger-value-oriented quality services is crucial for public transit companies if they are to satisfy their passengers and thus increase re-patronage/word-of-mouth behavior, and consequently customer loyalty.

Amal Hosny, Hi, If I understand your question, I believe you are conflating two issues. It's possible to have two ranked variables which clearly differ in, or don't differ in, their median (both tend to similar low values, etc.)--and yet have the _correlation_ between the two variables be very weak. In your example, this could happen if it's true that the satisfaction answer tends, collectively to be Low and the FlyAgain answer tends, collectively, to be low, but for individuals, the factors for when these low values are assigned are totally or largely unrelated. Additional noise can be added for considerations of how and why people assign possible non-low values, such as 4 versus 5 versus 8, etc. . Tools like Mann-Whitney are generally for comparing separate groups--like how males answer satisfaction versus how females answer satisfaction. But you're looking at dependent variables for how the same individuals answer a pair of questions.

If you are interested to study PHD program in Thailand , you have to contact to Professor of that Univ. or Institute working in the field related to your research, Take recommendation letter from him, then write to univ. / institute for PDF fellowship.Unless you have fellowship , it is difficult to pursue the said programme.

# Pantouvakis, A. and Patsiouras, C. (2016), "Exploring the role of leadership style on the service quality-customer satisfaction link: Evidence from a B2B environment", International Journal of Quality and Service Sciences, Vol. 8 No. 1, pp. 88-101. https://doi.org/10.1108/IJQSS-01-2015-0006

# EMPIRICAL EVIDENCE FOR THE RELATION BETWEEN CUSTOMER SATISFACTION AND BUSINESS PERFORMANCE. file:///C:/Users/USER/Downloads/erimrs20010613144003.pdf

Following

It depends on many things: objectives, study design, data collection methods, distribution etc. If it's a cross sectional study with normal distributed data then you can use easier formula used commonly. Your topic is basically qualitative one. Anyway you focused on quantitative approach and you do by using scales but should be used mixed method for getting better outcomes and also in-depth in this type of topic.

You need to design items to measure the constructs (customer satisfaction, fulfillment and assortment) that you have. How do you get price data from a survey? Is it over time or for 5 years? Survey data is at a pt. in time unless you want to do longitudinal data which is not easy to survey.

... This study aims to show the impact of the logistics management on customer satisfaction in small and mid-sized Algerian industrial companies, by interviewing the companies managers and everybody in charge of the logistic process, the interview questions will be based on some literature review issues. ...

Alexander, K. (2013). Facilities management: theory and practice. Routledge.

Totally agree that you need to keep the questionnaire as short and concise as you can. (And on re-reading my response not entirely sure I am being helpful in regard to length.) However, it would seem that you have not analyzed / broken down the constructs as thoroughly as you could. i.e. different variables make up the constructs.

Early in the questionnaire you should test awareness of brands … maybe it is a simple factor that people are not aware of the product/brand.

You may also be interested in the pricing differentials. i.e. if brand X is the premium brand you could look at how much cheaper the closest competitor has to be to switch purchase. Maybe brand image is not as important as price!

Then presuming you have established the buying factors either by groups or a literature review, you would position the different brands on the scales that describe them. (e.g. for a food product the two ends of a semantic differential could be extremely pleasant taste / not at all pleasant tasting)

If I was measuring brand loyalty, I would consider several indicators and where applicable have a probe: Why do you say that?

(In the following I am trying to think of the type of questions that would be useful. Assume brand X is the focus…)

1) Purchase profile:

Which brands of product are you aware of? (spontaneous)

Show list Which of these brands of product have you ever used?

If uses target brand: How many out of the last 10 times that you bought this product did you buy brand X ?... or could rather ask this Considering this list of Brands – of your last 10 purchases of this product how many were Brand X, Brand Y, Brand Z and so on.)

2) Attitude to brands: For the top three brands: What words would you use to describe brand X? Brand Y? Brand Z?

Is Brand X better/the same/ not as good as other brands?

3) Usage time: For approximately how long have you used Brand X? OR When did you start using brand X?

4) Substitution: If Brand X was not available would you

a) Not buy the product at all

b) Go to an outlet where you could buy brand X

c) Buy another brand

d) Buy another product type

e) Other (state)

5) Recommendation: How likely are you to recommend brand X to your friends? (very likely / likely / neither likely, nor unlikely / unlikely / very unlikely)?

6) Unique benefits: What benefits does brand X have that the other brands do not have?

7) Comparison of brand features --- Use Likert scales to measure the importance of product features. Then use semantic differentials to get the respondents to rate the different brands on the product features.

You can see that in some of the above one question can provide insight into different aspects of the various brands relative positions….

Hennig-Thurau, T., & Hansen, U. (Eds.). (2013). Relationship marketing: Gaining competitive advantage through customer satisfaction and customer retention. Springer Science & Business Media.

The value perceived by the customer is essentially the added value that the product adds to the competition ...

To have equal sample sizes is not at all a must.

Dear Vladan Pantelic ,

If you're seeking for literature related to this topic, the following papers might be helpful:

Extensive research on this topic has been done. If u still want to pursue this topic. Search for new variables.

To measure performance, I would not worry about the sample. If I would worry with whom to compare me. I understand that the key is to locate the best in the world, the best regions of the world in the case of the issue and other actions that are culturally very close. The comparative evaluation is good, it will show where the disadvantages are, analyzing costs, tariffs, inputs, resources, quality of processes and losses of all kinds. Take the opportunity to share a publication in which you will compile various cases of the public sector.

Thank you! Following! Best regards: Julia Doncheva

Anything you can to us Dennis Mazur ?

Interesting topic. Scientific research in specific fields of art, such as theater, for example, is now particularly important because the development of theatrical art is determined by many factors. On the one hand, the development of the theater is threatened by new online media. On the other hand, in some countries, along with the increase in citizens' incomes, the number of people visiting theaters is growing.

Best wishes

You can read the end of my JBR (2014) :

SERVQUAL cannot really 'measure' service quality. Quality of service is subjective, determined by the receiver. Therefor its measurement should be based on a measure of customer perception. A 10.000 feet definition would be Q=E-P, quality of service is expected service minus perceived service. Determine key result indicators for both, and do the math. A negative outcome reflects a negative perception, etc.

For the concept of satisfaction, you must first do a research in a theoretical framework: The Dissonance Theory, The Contrast Theory, Expectancy-Disconfirmation Paradigm, Inferred versus Direct Disconfirmation, The Comparison Level Theory, The Value Percept Theory, The Equity Theory, The Evaluative Congruity Theory... Then you propose the model of your research. Regarding the elements of the marketing mix, it's much easier.

here are some works that can help you:

Bearden, W.O. & Teel, E.J. (1983). Selected Determinants of Consumer Satisfaction and Complaint Reports, Journal of Marketing Research, 20, 21-28.

Botterill, T. D. (1987). Dissatisfaction with a construction of Satisfaction, Annals of Tourism Research 14: 139-141.

Cadotte, E. R., Woodruff, R. B ., & Jenkins, R. J. (1987). Expectations and Norms in Models of Consumer Satisfaction Journal of Marketing Research 24, 305-14.

Erevelles, S. & Leavitt, C. (1992). A Comparison of Current Models of Consumer Satisfaction / Dissatisfaction Journal of Consumer Satisfaction /Dissatisfaction and Complaining Behavior, 5, 104-114.

LaTour, S. T. & Peat, N. C. (1979). Conceptual and Methodological issues in consumer satisfaction research, Advances in Consumer Research, 6, 431-437.

Tse, D. K., & Wilton, P. C. (1988). Models of Consumer Satisfaction Formation: An Extension. Journal of Marketing Research, 25(2), 204. doi:10.2307/3172652

Olander, F., (1979) Consumer Satisfaction: A Sceptic’s view. Aarhus Denmark.

Oliver, L. R. (1997). Satisfaction a behavioral perspective on the consumer, The McGraw-Hill Companies, Inc. New York.

Model of operational risk management based on the synergy of quality management principles, risk mamagement principles and total quality management approach.

Please check the article in pdf. format available at:

As the sample size is robust two-way ANOVA can be used. But for large data it is important to look at histograms to understand the summary measures. The outliers also need to be addressed. Post this data editing regression will deliver results if the dependent variable is on metric scale.

You definitely can. You could either develop independent models for each aspect of the inspection process, or develop a single multi label model for the whole inspection process.

I'd probably consider only using ml / ai models for characterising ripeness and uniformity and using other computer vision techniques for characterising colour.

In terms of real time operation, something like the Nvidia Jetson tx2 board would allow you to run complex machine learning algorithms in real time and meet the time constraints you specify.

Hope this helps.

Alex

To quote CK Prahlad from his foreword to the book (Employees First Customers Second), this book is a narrative of organizational transformation and a practical guide for managers who wish to achieve similar results, as well as a story of the personal transformation of a CEO.

What struck me as significant about the book were the following aspects:

It is about the journey of an organization very Indian in its origins and culture

and about the experiences of an Indian CEO in bringing about change within

this psyche – this means it is a journey we can all relate very well to and draw

a lot of practical lessons from.

One small note to my previous post:

when measuring customer satisfaction, it is sometimes appropriate to consider the "zone of tolerance" – The difference between level of quality expectations and level of quality desires.

See for example:

Van Riel, A., Semeijn, J., & Janssen, W. (2003). E-service quality expectations: A case study. Total Quality Management & Business Excellence, 14(4), 437–450. doi:10.1080/1478336032000047255

I don't understand the question. A model doesn't come into being like a recipe that you adapt as you like. If you want to work empirically (that's how I understand your question), it has to be driven by theory, right?