Operated by the National Physical

¶ … operated by the National Physical Laboratory (NPL). The questionnaire is available online at http://resource.npl.co.uk/customer_satisfaction/questionnaire.html. The online nature of the questionnaire means that all customers may have access to completing it and it is not necessary to be invited to do so. This questionnaire was accessed by one of the group members after receiving an email invitation which directed them to this site. The email was received by the group member shortly after having used the services of NPL and was received as a thank you for using the services of NPL.

The purpose of the questionnaire is to collect information from the company's clients which specifically relates to the quality of the service which is provided by them. This is evident in the style of the questions which are asked and is also specifically outlined in the information which precedes the questionnaire itself. The questionnaire aims to collect information from all clients who have used the company, although one section is aimed specifically at those customers which had used a specific service. This is clearly detailed on the appropriate section of the survey however. It would appear from the information given on the Web site of NPL that this is the only element of the company's sampling strategy to gain information on quality of service. This may also explain why the company is eager to gain the opinions of all customers through this survey.

As the questionnaire is accessible to all it would indicate that the NPL wishes to target all customers as the population for their survey. This would also appear to be true given that all customers seem to be directed to the survey through email directly after receiving their order. This is important as it should enable the company to receive a very accurate representation of their customer's opinions by encouraging more of their customers to participate in giving feedback (Kaye & Johnson, 1999). As the survey is automated the large numbers of responses which are generated should not cause the same problems as high volumes of paper questionnaires may create. This is because automation should also allow for automatic analysis of the questionnaire results rather than manual input and coding in a statistical package. This should also allow for any bias associated with customer selection to be eliminated. For example if only a select number of customers were to be selected randomly to complete the survey this may result in regular customers having a higher probability of selection. The fact that these customers are repeat customers may in itself indicate that the customers are happy with the company's service, which may positively bias the results.

Overall the questionnaire is well designed. It is possible that the online nature of the questionnaire would encourage more to participate than a paper-based survey which requires more time to post back to the company (Kaplowitz, Hadlock & Levine, 2004). The language which is used in the questionnaire makes it easy to understand and it is always clear which areas of questions a particular respondent should answer. This is important as the style of questions has been shown to be influential in participation rates in Internet-based surveys (Dillman & Christian, 2005). The length of the questionnaire is also good, as enough questions are included to allow for a detailed analysis, but not so many that the respondent is inconvenienced (Trouteaud, 2004). Information is collected at the beginning of the questionnaire which allows for an understanding of whether the company completing the questionnaire is a regular customer. This may help to establish whether responses represent an overview of a number of experiences or one specific occasion. In addition, information is collected relating to the specific services which were purchased by the company. This may allow for an analysis of whether any reported problems are isolated to one specific sector of the company. This is useful at a later stage when solutions to the problems are planned and implemented. This may also indicate whether there are some elements of the survey further down which are more relevant for some respondents than others. For example if a respondent indicates that they used NPL for training or consultancy this may indicate that they would require regular contact with the company. This may then indicate that the areas of customer service which relate to contact with the company are particularly important for these customers. If however the customer indicates that they used the company for testing purposes then this may indicate that the elements relating to timeliness of project completion may instead be more important to that respondent. By collecting this information it may be possible at a later stage to analyse the information for each individual customer set to allow for improvements in specific business areas.

The largest section of the questionnaire is that which focuses on collecting information specifically about quality of service. These questions use predominantly a five point scale where individuals indicate how they judge each element of service. The scale is clear on the questionnaire, with 5 representing excellent and 1 indicating poor. The use of a scale is useful as it allows for numerical analysis to be performed which eases the comparison of different sets of data, for example comparison of surveys taken annually (Miller & Salkind, 2002). In addition to this, where the specific elements of service are addressed in the survey there is a 'not applicable' option. This further strengthens this area of the survey as without this option respondents may simply not complete certain questions. It may then not be clear as to why the customer has chosen not to complete these questions and in some cases may lead the researcher to unnecessarily discard incomplete questionnaires.

Another good element of the survey is the way in which it first asks for a general overview of service quality, value for money and ranges of services offered. This allows for a better idea to be gained of the general opinion of the company before asking for opinions on specific elements of service. By then further probing specific areas the NPL would then be able to pinpoint the specific areas in which improvements may be made. Furthermore, the questionnaire specifically asks the respondent to indicate the elements which are considered to be most important. This is useful as it allows for further prioritisation of the areas of service which require attention.

There are some elements of the questionnaire which may however possibly be improved. The first of these lies in the requirement to provide identifying details. The survey asks for details including the company name and reference number, and also asks specifically for the name and job title of the person completing the questionnaire. It is possible that providing these identifying details may deter some from completing the questionnaire (Joinson & Reips, 2007). This may be particularly true if the person wishes to indicate negative responses, especially if the customer is a regular user of the company. For example the individual may either choose not to complete the questionnaire, or they may choose to give positive responses to questions in which they would give negative answers if there were no fear of reprisal. The implications of this in interpreting the survey are significant as it may introduce a bias to the results which shows more positive responses than would occur in a random sample of anonymous users. It may help with the interpretation of the results of the survey if the job position of the individual is available to researchers as this may help them to establish the accuracy of the answers provided. It would not however be considered important to be aware of the name of the individual completing the questionnaire, or even the company for which they work. Instead, the importance associated with the answers could be judged solely from the information given as to how frequently the service is used, and the details about the activities in which the company engages.

There are also some questions which are asked which may be useful to the company for other reasons, but are not necessarily related specifically to the quality of service provided by the company. For example the sixth question on the survey asks for details of the reasons for which the NPL was chosen by the individual or company. Although this may give a vague impression of whether business has resulted from recommendation, it does not specifically ask this question. Therefore the question may be more useful to the company in determining their major business channels rather than illustrating anything specifically related to quality of service provided by the company.

Part B

The questionnaire which is detailed in this section is designed to collect information from customers of a photographic studio which is a franchise branch of a national brand. The customers who would be surveyed would be those who purchased one of numerous available packages to visit the studio for a photo shoot. The studio in which the questionnaire would be used is one in which one of the group members has previously worked and has access to conduct the survey. This means that the questionnaire may be designed with a thorough understanding of the processes within the company and the experience to which the customers of the company should be ideally exposed.

The studio presently collects some information on the experience of their customers but this focuses predominantly on the customer's experiences on the day of their shoot. For example the questionnaire which is currently distributed to all customers asks questions such as 'Which parts of your photo shoot did you like the best?' The current questionnaire is also very simple with only three options given for the multiple choice questions and a number of questions such as that above which are open ended questions. The company is however concerned that sufficient information is not available to guide them specifically on customer satisfaction with the service provided from initial contact with the company to receipt of the photographic products ordered. The survey is therefore focused solely on the area of service received rather than the opinions of the customer which relate to the photo shoot. The objective of collecting this information is to identify whether there are any problems relating to customer service alone. If there are any problems identified this may then allow the manager of the studio to design and implement changes which allow these problems to be resolved. This is likely to be useful to the manager as better customer service is likely to further increase overall customer satisfaction. This is particularly important for the manager as recommendation by friends has previously been found by the company to be the most important marketing channel. Therefore improving customer service may lead to substantially increased revenue for the company.

The nature of the company meant that there were several different stages which may be reached by the customer, not all completing the entire process from beginning to end. For example the sale of gift vouchers meant that there may be one customer who first contacted the company but that they would not then undergo the actual photographic session. Equally, there were often families in which a large number attended for the photographic session but not all would then undergo the selection and ordering process. There were also often family members who may arrive at the studio for ordering of products but who may have no other contact with the company. Therefore the different levels of contact with the company needed to be considered when determining which customers' opinions would be most useful to the company. After consideration and discussion with the manager of the studio it was determined that all customers' opinions would be valued for the purpose of determining service quality. The justification for this was that recommendation of the company could come from sources who had much or little contact with the company, but no matter how small the contact, judgement may still be made based on that contact. Therefore it was determined that all customers of the company would be included in the sampling population.

Both the objective of the study and the decision to include all customers determined the information which needed to be collected by the survey. A copy of the first draft of the survey may be seen in Appendix 2. It was determined that the survey should remain anonymous as it was felt that this may encourage wider participation (Joinson & Reips, 2007). For example some customers who were intending making further visits to the studio may feel uncomfortable giving negative responses if they thought that there may be repercussions to their future experience. It was also determined to be unnecessary as it would have no bearing on the analysis of the survey.

As all customers were to be included in the questionnaire it was necessary to use the first question to collect some information about the extent of the contact which the respondent had with the studio. The first question therefore asked the respondent to tick next to the statements which were applicable. This also allowed for information to be gained about the specific stages of the process which the respondent had undergone. This not only helped to analyse how much contact the opinion was based upon but also helped to identify the team members from whom the client had contact. This would then allow for more in depth analysis which could be based on specific elements of the entire studio experience. For example respondents whose only contact with the studio was to purchase the package could be isolated. Analysing the responses from these customers would then specifically identify problems with the quality of service provided by the sales staff at the studio. This would then make it easier to identify the specific areas in which improvements could be made, thus making the survey more useful to the manager than if no such information were available.

The main emphasis of the survey was on the second question, as this was the measure of the different elements of service which the respondent had received. A five point scale was used for the answers as this would allow for statistical analysis of the answers. Quantitative results would also allow for an easier comparison between results collected at different times (Miller & Salkind, 2002). The different elements which were assessed in the second question of the survey were identified as the most important elements of service which were commonly experienced by customers. It was recognised that some customers would not have sufficient experience in these areas to comment however and so there was also a 'not applicable' option included.

The third section of the survey focused solely on the satisfaction of customers with complaints which they had made. This was isolated for further analysis as this is an area which the studio manager identified as having been a problem area. It is possible that this may also help to identify whether those customers who had made complaints to the studio were also those who had negative service quality experiences overall. It is then possible that separate analyses could be conducted of the respondents who had made complaints and those who had not. This may also help in the development of specific strategies to deal with customers who complain to the company. For example it may be possible that large numbers of customers who had made a complaint also identified that they were unhappy with the notification of progress. This would then indicate that by increasing contact with those who had complained it may be possible to improve their overall perceptions of service and compensate for the original problem.

The fourth question asked customers specifically whether they would recommend the studio to their friends and colleagues. Although this was not specifically related to the levels of service quality received it provides a good overall view of the respondents opinion of the service received. It also allows for analysis of this in relation to the specific objective of the survey, which is to determine how well service quality may be contributing to sales and how this may be improved. Finally the survey allows the respondent to openly add their own comments. It would not be expected that all customers would use this space or that many of the suggestions would be practicable. It is however possible that there may be further information which may be gained which was not anticipated in the original questionnaire design.

After the decision had been made to include all customers of the company in the sample population a test survey was conducted to determine whether this was practical and to test the survey designed. The company holds databases with the details of all customers with details taken upon initial contact with the company. One database includes contact details of every customer who purchased a package from the studio, while another database included details of every person attending a photographic session, and a third database containing details of all those who purchased photographic products. The initial step was therefore to accumulate all of these databases and then to remove any duplicate entries. As the survey was aimed at adults the names of all those under 18 were also removed from the database. The decision was also made that customers may less accurately recall their experiences if a long period of time had elapsed and so those who had no contact with the studio in the last six months were excluded from the database. This then resulted in one database with all those to be included in the target population.

From previous experience it was clear that around two thirds of all those attending the photographic session completed and returned surveys which they were given. It was therefore assumed that this rate of participation would be constant across all customers. The surveys which are currently given to customers are distributed to all visiting the studio for a photographic session. The distribution of the new survey to all customers would likely therefore result in large quantities of information which would be time consuming for the studio to analyse. Instead it would be advantageous to survey only a cross section of the customers. For the pilot survey it was therefore decided to send out 15 surveys in anticipation of 10 being completed and returned. In addition to the questions which were included on the survey detailed in Appendix 2 respondents were also asked whether they found the survey easy to understand and complete. This question was included so as to gain a better understanding of whether the survey was able to provide valid results.

There are a number of different sampling strategies which may be used for selecting participants for such a survey. Probability methods such as systematic sampling could be used as it is unlikely that if customers were listed in alphabetical order there would be any other underlying bias. Given that the list is so long and that there were only a small number of participants required there would be little difference in the results obtained from this or random sampling. Stratified sampling could also be used as this would ensure that a specific number of each different type of customer were selected for inclusion. It was however considered possible that customers given the survey may pass it to another member of the family who also attended the session and this would therefore disturb the strata used. Convenience sampling is an example of a nonprobability method which could be used (McQuarrie, 2005) and this may be useful in the pilot study, for example by giving the survey to the next 10 customers to visit the studio. It is however possible that this may provide a biased view as it would be based heavily on the service provided by staff at the studio on that particular day.

In the database all entries were automatically numbered in alphabetical order and Microsoft Excel was used to generate 15 random numbers. The entries in the survey which corresponded to these numbers were then selected for participation. To reduce postage those customers who were due to visit the studio at a future date were handed their survey on their next visit. All others chosen were posted their survey along with a return envelope addressed to the studio.

In response to the pilot survey there were 11 respondents who returned their questionnaire. This would therefore indicate that the response rate to a wider survey would be around two thirds of those selected, which is as originally anticipated. Overall there was a positive response to the survey, with all participants indicating that they found the questionnaire easy to understand and complete. In addition, all questionnaires were returned fully completed as appropriate which further supports this assumption. Although it was anticipated that a small number of respondents would offer additional suggestions there were none in the pilot study who completed the fifth question. Despite this it was deemed worthwhile to keep this section on the questionnaire as there may still be some who would wish to utilise the opportunity to offer suggestions. Therefore it was decided that the first draft of the questionnaire design would also be suitable to use as the final version within the studio if the survey were to be extended.

The results which were obtained from the pilot study also illustrate how the survey may be useful to the studio manager. The results from the survey were input to Microsoft Excel for analysis and there were found to be some trends in the information even though the surveys reflected the opinions of only a small sample of customers. The chart in Figure 1 shows the proportions of customers which would recommend the studio to a friend. From this chart it is clear that there are no customers who would definitely not recommend the studio, which indicates to the manager that there may be no problems with their reputation at the present time. However the number of those who would definitely recommend the studio varies according to whether they actually came to the photographic session themselves or not. This would indicate to the manager that there may be some area of service delivery which may be improved upon for those not attending the photographic session. The answers from question 2 on the surveys of those customers may indicate the specific areas which may be improved.

Figure 1: The responses on the survey to the question "Would you recommend the studio to a friend or colleague?" according to whether the customer bought the session for themselves or to give as a gift, or if they received the package from someone else

Part C

Using the data taken from the customers it is possible to construct a time series analysis to observe whether there are any trends present in the customer satisfaction ratings given to the company. The raw data was therefore used to plot the value of the Customer Quality Satisfaction Index against each month over the four-year period. The results of this time series analysis may be seen in Figure 2.

From this figure it is possible to observe that there is a relatively steady score achieved over the first two years of the survey for customer satisfaction. Over the last two years there appears however to be a clear upward trend. This upward trend contains some peaks and troughs but these do not appear to be seasonal in nature. It is difficult to speculate whether these variations are random in nature or due to some particular event within the company. For example there appears to be a slight decrease in ratings between February and June 2005, which may be due to a particular problem which was being experienced by the company over this period. It is however entirely possible that this was merely due to random variation, and does not appear to affect the overall upward trend in ratings. Given that the overall trend is not particularly affected by this variation it will therefore be assumed that it is random in nature. It is therefore then apparent that while the customer satisfaction levels for the company were consistent between August 2002 and July 2004 they have been consistently increasing between August 2004 and July 2006.

Figure 2: The monthly Customer Quality Satisfaction Index for customer satisfaction using data taken from 1000 customers each month between August 2002 and July 2006

Using the time series analysis which is shown in Figure 2 it is also possible to create a forecast of whether the increasing trend may continue within the near future. The most accurate method to predict the value of the Customer Quality Satisfaction Index would be to use the exponential smoothing method. This would involve using averages of the values which are already known and take into account the increasing trend to predict the most likely value of the index in August 2006. As the forecast which is required is for the next six months and not only one month it is not appropriate to this method however, as it is able to produce a future forecast for only one month. Therefore regression modelling must instead be used to create the forecast (Shim & Siegel, 2006). As with the production of the time series analysis which is detailed above, Microsoft Excel was used in order to produce the forecasts and models detailed below. This program automatically produces many of the calculations, which means that these will be minimised within the discussion.