Service Failure & Recovery Research Project

Part 2

MKTG 3710.001

MD ROKONUZZAMAN

FALL 2016        

“The Eagles”

Allison Davis

Brianna DeLashaw

Elizabeth Cannaday

Erin Strickler

Marlan Franklin

Table of Contents                        

Executive Summary                                                                        3

Summary                                                                                4

Research Objectives/Questions                                                        10

Methodology and Data                                                                        11

Data Analysis and Findings                                                                14

Conclusions and Recommendations                                                        18

Limitations of Research                                                                19

References                                                                                20

Appendix                                                                                21

Executive Summary

The research conducted by our group dealt with service failure and recovery and how it can influence the relationships between businesses and consumers. A service failure occurs when a company fails to meet the expectations of the customer. When those expectations aren’t met, often a negative experience for the customer follows, leaving them unsatisfied and unlikely to conduct business with the company in the future.

We began our exploration of this issue with introductory research to determine what came to the minds of consumers when asked about service failure and recovery. Next, we surveyed consumers to find what specific factors are important to them when dealing with a service error. Our group used several different distribution methods to survey approximately 175 people. The data was then recorded, organized, and analyzed in an effort to understand the emotions felt by customers, what efforts should take place after customer expectations aren’t fulfilled, and what constitutes these efforts as successful or unsuccessful.

Following the collection of data through surveys, our group used the responses to run statistical analyses such as:

• descriptive statistics on demographic variables
• independent sample t-test
• analysis of variance (ANOVA)
• Chi-square test
• bivariate correlation
• multiple regression

Along with the results of these analyses we will consider any limitations and problems that arose with each test that could potentially cause misrepresentation of data. The information we have gathered can help businesses better understand consumers and their perspectives - an integral approach to ensure positive future interactions between businesses and their customers.

Summary

Brief Introduction

Service failure is an issue that almost all industries will experience at some point in time, especially in industries that are service dominant versus industries that are product dominant. There are countless ways a service could fail consumers. In most cases these failures leave the  customer with a negative opinion on the company, especially if there has been no attempt made to resolve the issue. After the failure has taken place, losing current and potential business from an unsatisfied consumer is a relevant concern to have. Most companies will attempt to gain back consumer preference through some form of compensation.

Another issue companies may face is consumers attempting to take advantage of a company by claiming they were wronged even if the expected level of service was met, or the quality of the product was adequate. There is a certain level of difficulty in identifying the line between those trying to take advantage of the company and those that were genuinely failed by their service, all the while trying to correct problems for future consumers.

Objectives

The objective of this research is to describe how service failure and recovery can have an impact on the relationships between consumers and businesses. The information we discovered through conducting surveys can help understand what kind of emotions are experienced by consumers when a company fails to meet their expectations. The focus then moves to understanding what actions a consumer might take in relation to those emotions following a company’s service failure and recovery effort, or lack of recovery effort. With insight gained from this research to how consumers feel and react to service failure and recovery effort, businesses will be better equipped to make decisions that promote positive interactions with customers that will continue to happen in the future.

Background

The problem we are facing is that there are multiple cases where customers have received poor, and sometimes intolerable, service from companies or professionals. The real issue is not technically the service failure itself, but more importantly, the ability to recover in a timely and efficient manner. Service failure lacking recovery is a major issue, in that it can lead to a decrease in customer loyalty, the spread of negative word of mouth to current and prospective customers, and even a bad reputation for the company. It is essential to have strong service recovery because if it is absent it can lead to a major drop in company profits. It is also important for customers to know they are being taken care of when exchanging currency for a particular product or service. There are many scenarios where service failure can become evident, such as lack of communication, dishonesty, poor attitude or behavior, and not following up. Consumers expect every company to provide the most excellent service possible, but they cannot be perfect, so it will not always delivered. The main purpose of this research is to observe cases of service failure in depth, and observe how that company was able to recover from their poor service. This research also is to observe the customer’s reaction and overall attitude toward the company once the situation has or has not been resolved.

Analysis

The research reveals that when businesses fail to meet customer’s expectations it can greatly affect their perception of the organization as a whole. It leaves them with negative feelings such as dissatisfaction, annoyance, betrayal and usually leaves them stressed (refer to appendix). The initial mistake made by the company hurts the consumer’s perception of them, but the way they redeem their error is what leaves a lasting impression. There is such thing as a Recovery Paradox which suggests that “customers who are dissatisfied, but experience a high level of excellent service recovery, may be more satisfied and more likely to repurchase than are those who are satisfied at the first place” (MBA Knowledge Base, 1). For example, a teenager who’s video game account had been hacked, contacted the support channel. Not only had the issue resolved in a shorter amount of time than promised, but the customer service representative provided details on how to get in contact with them for help in a shorter amount of time. Because the “recovery exceeded his expectation,” this teenager will continue to be a loyal customer to the company (Stevens, 2).

On the contrary, in an in-person interview conducted during the research process, a respondent was asked how they felt when a company carried out a successful recovery effort versus an unsuccessful one. The respondent answered, “..that determines if I’m going to do business with them again or not. I don’t forget easily when it comes to companies that have done me wrong in the past or that I’ve bought products from in the past that were not quality products. If the recovery process isn’t outstanding then I generally don’t even worry about doing business with them again in the future. So it’s extremely important.” It is a given that not all companies can be perfect, and that employees may make mistakes every now and then, however, it is still an inconvenience if the service failure occurs often and is not recovered in a timely and sufficient manner. Thus, it is beneficial for the company to understand the negative consequences that can follow a service failure and how much it can affect the consumer's perception of their company in the future if they do not correct it. It is in the best interest for companies to have a plan of action for not only how they are going to recover after service failure but they should also have a plan to help minimize failures as a whole. The recovery process is key to a company’s success because the first service failure a customer experiences with a company is more prominent in their minds than any errors that follow. Therefore the company wants to make sure that they correct any mistake they made in order to prevent losing future business with the consumer or even taking the chance of the consumer spreading negative word-of-mouth about their company. So the way that they then choose to recover from the mistake is very important in terms of how the customer will remember them after.

Human error tends to be common in organizations. No matter what industry, it can be assumed that an employee will make a mistake or even fail to handle it according to the customer’s expectations. It is difficult, maybe even impossible, to know when these errors will occur. Since the timing and frequency of service errors are unpredictable, it is important to focus on error management. Error management is a strategy that focuses on minimizing the negative consequences of errors by early detection, quick error correction, and on preventing similar errors in the future by analyzing the causes of and learning from errors (McCollough and Berry, 380). A great example of error management was found during another in-person interview we conducted with a nurse practitioner working in the stroke/neuro unit of the Medical Center of McKinney. The very nature of the healthcare industry causes potential for a lot of the errors to occur every day. Since people’s lives are at risk there is high pressure due to time limits, and the amount of people working long hours can cause fatigue and therefore careless errors can often take place. During this interview it was stated that branches within hospitals often have meetings where they discuss how the branch as a whole is doing, what problems they are having, what they are doing to prevent these problems and problems similar to them, and what they can do to recover from any errors made.

Once the organization has made an effort toward error prevention, they are already on the right track to managing good business to customer relationships. But in the event that a service failure does happen, the next step is to decide what actions need to be made to recover from the error in an effort to maintain loyalty from customers. Research has shown that by simply having a conversation with the customer who has experienced a service failure is a major step to recovery. The conversation should be customer oriented, meaning the needs of the customer should be heard and considered. During the same previously mentioned interview, the nurse practitioner stated, “I would definitely say that listening is a major, major key to success with customer service failure. And by listening I mean giving them my full-undivided attention. You definitely want to let people know that their problems are being heard. I always just basically let people vent to me for a little while because that not only helps them get it off their chest, but I need to know everything that they feel they aren’t getting so I can try to provide that for them the best possible solution I can come up with.”

It is important for any company in the service industry to understand the profits gained and lost throughout the service failure and recovery process. The losses include “money-back guarantees, warranty work, or replacements” which fall back onto the company (Hart, 6). If the recovery is done successfully, eventually this will not be considered a debt for the company, but will turn into profit. In a research study for Club Med, the numbers showed that losing a customer was more expensive than a successful recovery effort. This proves that it pays to recover the right way, no matter that costs at hand in the beginning.

Another key factor in the service failure and recovery process is the platform upon which the consumer expresses their issues. Whether it be some form of social media, through email, through a call center, or just word-of-mouth, this can affect the way others see the company after the issue is resolved (see appendix). A huge, and still growing, aspect of the decade is social media. Through this, people can post to all of their friends and family to complain about a terrible experience that they had with a company or service. This spreads negative word-of-mouth like a wildfire. While some companies have joined the social media world and attempted to recover by responding directly to the dissatisfied consumer, most will usually not see this and all of the people that see the post will be left with a negative perception on this company. For example, in an in-person interview, a respondent explained that American Airlines direct messaged them after seeing a Facebook post about a service failure with their company. Although the recovery was successful, anyone that saw the Facebook post most likely did not see the direct message attempting to regain the respondents loyalty, and still see American Airlines as a company that failed to provide a satisfactory service.

Conclusion

In conclusion, it was found that consumers are understandably disappointed when they are not provided with the service or product that they were promised. In a lot of cases it leaves the consumer with a negative perception on that particular company or industry. It may even lead them to stop interacting with the company entirely or to talk poorly about that company if they are ever brought up in conversation. Consequently, the recovery process is extremely important to not only the consumer, but to the company as well, especially if the company wants the opportunity to continue exchanges with that particular customer.

Research Objectives/Questions

The purpose of this report is to identify what causes service failure and what actions can be taken to recover the situation so that the customer is content and will continue to use that company. Our key objective is to discover the likelihood of a customer to continue business with a company after a service failure. The first question whether or not the employee reacted promptly to the situation. They could have either claimed that they “agree” or “disagree” or somewhere in between. If they agreed, as opposed to disagree, then the rest of the survey was more meaningful and has more depth. The more in depth the survey is, the more our research can discover what makes customers choose to continue to do business with that company and their overall attitudes of the experience.

Our next question was if the employee attended the problem quickly. This will help us obtain a certain idea, as far as the effects of a fast recovery. If they agreed, then it is possible the survey will lean towards an overall positive experience in that given situation. If the respondent had disagreed, then the only true result is that it was an unpleasant experience and they will most likely never do business with that company again.

Although the first few questions refer to the speediness of the service failure recovery, our other objective is to discover if the employee personally corrected the situation. The next few questions ask whether or not the employee apologized or showed any regret of the situation. Depending on the response, there is so much depth in these questions, considering the customer-employee relationship is currently being observed. If the respondent agrees, we can ask more questions on how the employee was not only able to correct the given situation, but also if they were able to make the customer content and feel good about the given situation. These questions are most important because certain responses can lead to more in depth questions detailing the after of the service failure.                

Methodology and Data

                Our survey (refer to appendix) assisted in the gathering of data for service failure and recovery experiences.

Participants:

There were 125 surveys distributed randomly. 70 of the respondents were female, while only 55 were male. The majority of respondents are no longer students and are between 21 and 25 years old.

Measure:

        The measure used to conduct this survey consisted of the Qualtrics online survey tool and hard copy surveys. Respondents were asked to answer 60 questions ranking their experience with a service failure and recovery on a Likert scale (ranging from strongly disagree to strongly agree). The last 7 questions were asked to record demographic information to enhance the research process.

Procedure:

        As mentioned earlier, the surveys were distributed to 125 respondents randomly, 100 sent through Qualtrics and the remaining 25 through hard copy surveys. Each group member was required to send out 15 online surveys and 5 hard copies, although one group member received an additional 25 through the online segment. Participants were given a brief overview of the survey and were urged to answer questions to the best of their ability. They were also made aware of the confidentiality of the survey results.

Issues:

Repetitive

                Throughout the survey there were statements that could be deemed as ‘too similar’ to the respondent. For example, the first three statements in section A, referring to the quickness and promptness to the situation, can be tedious and annoying for the respondent to answer. This can also result in the second guessing of answers. Another example stems from section A in which the respondent might think that the adjectives “friendly,” “polite,” “respectful,” and “patient” have the same meaning. This can be resolved easily by choosing questions wisely and refraining from using repetitive words to avoid confusion and annoyance.

Misunderstanding

At first glance at the topic of “Service Failure and Recovery,” it might not make sense to an outside eye. This topic requires an explanation before the survey can be completed accurately. Based upon the introduction of the topic, the surveyor could have explained it differently, thus resulting in different interpretations among respondents.

For certain questions, such as 18 and 19 in section A, the statements should be ranked on a scale with labels other than “strongly disagree” to “strongly agree”. In the mind of the respondent, it would be more helpful to rank a question concerning “providing solutions” on a scale from “none” to “most”. This will give the opportunity for the respondent to answer accurately and fairly.

Presumption of Knowledge

A presumption of respondent knowledge can strongly affect the answers given. It was inappropriate for the surveyor to assume that the respondents knew what “procedures” the chosen company is required to follow in the given situation. Unless it is stated before the situation is handled, then the consumer will have no knowledge of what is the right and wrong procedure. This occurs in section C for questions 7 and 8 when asked about the company procedures being adapted.

Falsehood

It is extremely important in a survey to not ask generalized questions. Unfortunately, section F is saturated with questions asking about personal actions and emotions. It would be more beneficial to make statements about a specific crisis in order to get more accurate and detailed responses from respondents.

        Another aspect of falsehood can be described as lying or lack of effort or thought being put into the survey. There could be a small percentage of people that filled out the survey without actually reading any of the statements just to finish it. Also, if a respondent feels pressured, they could lie to alter their actual answers in fear that the surveyor is watching or just because they want to. Some people may subconsciously lie to make their story more entertaining or different than the reality.

Constructive:

Demographics

        It was incredibly beneficial including demographic questions to conclude the survey. The questions gave us the opportunity to link gender, age, income, etc. to the satisfaction of the service failure and recovery process.

Results

        The statements made in section E were the most effective out of the entire survey. The statements asking whether or not the individual would return or refer anyone to the company gives us as researchers a gauge of effectiveness of the recovery.

Data Analysis and Findings

Independent T-Test

The research team hypothesized that there would not be a significant difference between males and females speaking positively about the organization to others.

• Null Hypothesis (Ho): There is no difference between males and females speaking positively about the organization to others.

• Alternative Hypothesis (Ha): There is a difference between males and females speaking positively about the organization to others.

In order to test the hypothesis an independent t-test had to be used. Once the test was performed (see Table 1), we were able to see a Z score (or t score) of 1.137, suggesting that we have to accept our null hypothesis because the value falls within our acceptance range of ±1.96. To confirm this, we looked at the outputs to find our p-value. Since the p-value of 0.256 is larger than 0.05 we accept our null hypothesis. The conclusion of the test is that we accept our null hypothesis, realizing that there is no difference between males and females speaking positively about the organization to others.

ANOVA Test

The research team wanted to know if there was any difference in respondent’s loyalty to the organization between the different services. It was hypothesized that there was no difference.

• Null Hypothesis (Ho): There is no difference between the groups (services) in terms of being loyal to the organization.

• Alternative Hypothesis (Ha): At least one pair of the groups are different in terms of mean score.

To test for significant differences in loyalty to an organization among multiple services (our frequency group) we had to use the Analysis of Variance or ANOVA test. Based on the output (see Table 2), the p-value is 0.000. Since the p-value is smaller than 0.05, we know that at least one pair of services is significantly different. This is enough for us to reject our null hypothesis, but we need to look at the Post Hoc table to determine which pair makes the p-value significant. The following showed significant differences in mean loyalty to the organization:

• Other vs. Airport – Significant
• Other vs. Pet Clinic – Significant
• Other vs. Hospital/Clinic – Significant
• Other vs. Department Store – Significant
• Other vs. Restaurant – Significant
• Other vs. Gas Station – Significant
• Other vs. Hotel – Significant
• Other vs. Banking/Credit Card Service – Significant
• Holiday Resort vs. Restaurant – Significant
• Holiday Resort vs. Gas Station – Significant
• Holiday Resort vs. Hotel – Significant
• Holiday Resort vs. Banking/Credit Card Service – Significant

The conclusion of the test is that there is a significant difference in loyalty to an ‘Other’ organization and the loyalty to the organizations in all other service sectors besides the Airport.

Chi-Test

• Null Hypothesis (Ho): There is no association between household income and student status.

• Alternative Hypothesis (Ha): There is an association between household income and student status.

In order to see if there is an association between household income and student status we used the cross-tabulation test. The outputs (see Table 3), give us a p-value of 0.000, which is smaller than 0.05, meaning there is a significant association between the two variables. Since there is a significant association we reject the null hypothesis stating that there is no association between household income and student status.

Correlation Test

• Null Hypothesis (Ho): There is no correlation between being satisfied with the way the problem was handled, saying positive things about the organization if asked, and recommending the organization to a friend.

• Alternative Hypothesis (Ha): There is a correlation between being satisfied with the way the problem was handled, saying positive things about the organization if asked, and recommending the organization to a friend.

1. A Pearson correlation coefficient was output to understand what kind of relationship, if any, there is between being satisfied with the way the problem was handled, and saying positive things about the organization if asked. There was a strong, positive correlation between these two variables, r (1574) = 0.658 (see Table 4).

1. A Pearson correlation coefficient was output to understand what kind of relationship, if any, there is between being satisfied with the way the problem was handled, and recommending the organization to a friend. There was a strong, positive correlation between these two variables, r (1574) = 0.653 (see Table 4).

1. A Pearson correlation coefficient was output to understand what kind of relationship, if any, there is between saying positive things about the organization if asked, and recommending the organization to a friend. There was a strong, positive correlation between these two variables, r (1574) = 0. (see Table 4).

By meeting the requirements of having a p-value less than 0.05, and have strong, positive R-values (Pearson Correlation), we reject our null hypothesis stating that there is no correlation between the three variables.

Multiple Regression

• Null Hypothesis (Ho): There is no association between the time it took to solve the problem, and the level of satisfaction with how the problem was dealt and the continuation of getting services from an organization.
• Alternative Hypothesis (Ha): There is an association between the time it took to solve the problem, and the level of satisfaction with how the problem was dealt and the continuation of getting services from an organization.

In order to analyze more than two groups of nominal or ordinal data we had to use a regression analysis (see Table 5). The r-value tells us that 31% of the variance of not taking long to solve the problem is explained by these two variables.The p-value of both is 0.000 which is less than 0.05, therefore we reject our null hypothesis. This means that at least one pair of variables is significant. By next looking at the betas, 0.371 and 0.252, we know that there is a weak but positive relationship between our dependent variable and independent variables.

Conclusions and Recommendations

When further analyzing our test results we found that there was no difference between how males and females talked positively about the organizations to other. From this data we concluded that no matter the gender the way that people speak positively toward companies doesn’t not differ. Knowing this information, we think that it would be smart for companies to make sure that they have a good recovery process no matter the gender of the customer. This will ensure that the company will still be spoken about in a positive light by all consumers.

We then further analyzed our results using a different type of test and we found that there was a difference between the service in terms of being loyal to the organization. We found that at least one of the groups was significantly different. It would be beneficial for companies to build up customer loyalty by providing good loyalty programs.

We used a chi square test to test if there was an association between household income and student status. We were able to conclude that there was not an association between the two. This means that people of all incomes are students.

We did a correlation test and we were able to conclude that there was a correlation between being satisfied with the way the problem was handled and saying positive things about the organization to a friend. With that being said I think that it would be beneficial for organizations to ensure that they have a proper recovery process because the way the companies handle problems has a big effect of how their consumers talk about them afterwards.

Limitations of Research

        There were a few limitations to our research that indicated slight caution when identifying and interpreting aspects of the given data. We came across a few problems in the research with the surveys. Besides typical survey fatigue, misunderstanding of the question, and no responses, it would have helped our research to create a few more questions in depth in response to the respondent’s reply. Each respondent had an experience within a given department, whether medical, restaurant, or retail. It would have been more helpful to design questions in the survey based on the industry.

        For example if their service failure experience took place at a restaurant, a few questions to consider would be if they received any discount or complimentary items, such as a free meal or half off an appetizer. We could have also asked if the manager or someone in charge ever made an attempt to correct the given situation. If the service failure took place in a veterinary office, we could have asked if their pet received special care or services in response to the failure, or if the bill had applied discounts as well.

        Although we were able to conduct enough research to fortify our conclusions, it would have been more helpful if our questions were centered more around the industry as opposed to just being general about the situation. It would have also been more beneficial to have more questions based on the responses to the survey. This would have helped us gather even more information about the aspects of service failure and recovery.

References

McCollough, Michael A., and Leonard L. Berry And. "Michael A. McCollough." An Empirical Investigation of Customer Satisfaction after Service Failure and Recovery. N.p., 01 Nov. 2000. Web. 13 Oct. 2016. From

http://jsr.sagepub.com/content/3/2/121.short#cited-by

Hart, C. W., Heskett, J. L. & Sasser Jr., W. E. (1990). The profitable art of service recovery. Retrieved September 14, 2016, from https://hbr.org/1990/07/the-profitable-art-of-service-recovery

MBA Knowledge Base. Service failure and recovery. Retrieved September 14, 2016, from http://www.mbaknol.com/services-marketing/service-failure-and-recovery/

Stevens, B. (2010). Marketing examples: Service failure and service recovery. Retrieved September 14, 2016, from http://uwmktg301.blogspot.com/2010/03/service-failure-and-service-recovery.html

Appendix

Data Analysis Tables

Survey

University of North Texas
Survey on Understanding consumer perceptions about Service Failure & Recovery

Service failure refers to a situation when as a customer, you did not get what you expected. Such scenarios may include but not limited to ✓ Restaurant- You were served food that was not prepared according to your order or took too much time to serve
✓ Banking/ Credit Card service- Transaction failure
✓ Airport- When you arrived at the airport, you were informed that flight was cancelled

These are just some of the service failure examples. In addition to these scenarios, you may have faced other service failures not mentioned above.
Now, could you indicate where you faced such a SERVICE FAILURE incident that warranted a compliant? Please circle ONLY 1 and answer the following questions based on that particular experience -

1) Restaurant 2) Department store 3) Hotel 4) Banking/Credit Card Service 5) Holiday resort 6) Gas station 7) Hospital/Clinic 8) Pet Clinic 9) Airport 10) others [Please specify area] ---

A. Now please answer the following questions based on the service failure experience that you just selected

B. Now please think about the same service failure situation . What are your evaluations pertaining to the recovery steps provided by the Service Personnel?

C. . Please provide your opinion for the following statements regarding resolution of the same service failure situation:

University of North Texas
Survey on Understanding consumer perceptions about Service Failure & Recovery

D. Again, think about the same service failure situation and steps taken by the service staff. What are your feelings pertaining to the service recovery provided by the service personnel?

E. Keeping in mind the same organization, please provide your opinion for the following statements:

F. Now please answer some general questions about yourself.

Please tell us something about you:
1. Your gender? __Male __Female 2. Your age?

3 . Your Household Income?

4 You currently: __Do not work

__Work Part-time

__Own a business ____Divorced/separated

More than 3 __Not a student

5. 5  Your marital status?
6. 6  Do you have children?
7. 7  Are you a student?

___Married No 1 __Full-time

__Work Full-time

____Not married 2 3
__ Part-time

____Living together

We sincerely appreciate your participation in this important survey. Thank you for your time.