During an unusually slow period, I sent out a satisfaction questionnaire with the goal of bringing my services more in line with my clients’ requirements.
While wondering why my workflow had suddenly come to a halt during an unusually slow period in October 2011, I thought of a way to find out if my clients had been shopping around for cheaper or better language service providers elsewhere (without actually asking them the rather embarrassing question straight out). With the goal of offering services more in line with my clients’ requirements, I decided to send them a satisfaction questionnaire. The research I did to understand the thinking behind this kind of survey so that I could choose the right questions to ask gave me a totally new perspective on how my clients perceive the quality of my services. Although most of the replies to the survey questions were fairly predictable, some of the answers were perhaps a little surprising.
What I really wanted to accomplish by sending out this survey was to discover if my clients were still my clients, to find out if they had gone out of business, and to provide them with a reminder that I was still on the market. It was also a way of giving my working day a purpose, rather than just twiddling my thumbs until work arrived.
The true purpose of a client satisfaction survey is to see if your services are in line with your clients’ needs and to identify priorities for improvement. Satisfied clients typically become loyal clients. Finding out what satisfies them can help your business succeed.
I decided to include all of my Italian clients who had given me at least one translation job to do since January 1, 2009. Translation is my core business, amounting to 62% of my occupational earnings in 2010. Italian clients accounted for about 93% of my professional income in the same year. By excluding the non-Italian clients, the sample was more uniform and there was no need to translate the questionnaire into English. By examining other client satisfaction surveys available online, I was initially surprised to find that they are very often not anonymous. This is probably so that certain critical issues regarding only particular clients can be tackled in a specific way.
To save time, I checked online to see if there were any model or existing surveys I could adapt. I first looked for surveys performed by freelance translators based in Italy, but found none. Consequently, I widened my search to include freelance translators who had written their surveys in English, and still found none. At that point, I searched for translation agencies and came across a few surveys in Italian and several in English. However, upon closer analysis the kinds of questions translation agencies were asking proved to be inappropriate for freelance translators. For instance, many of the questions concerned the interaction between the client and the project manager. In general, I found that the type of questions asked by freelancers working in other professional fields were more appropriate to my situation.
Setting Up the Survey
There are many websites that give hints, tips, and advice on designing client satisfaction surveys. There are also several free articles and papers on the subject written by experts. Many websites provide survey hosting services, some of them free of charge. However, I chose to set up my own form written in Active Server Pages (ASP), with an underlying MS Access database.1
Since I was unable to find a suitable model to base myself on, I had to start almost from scratch by taking a basic theoretical approach, beginning with an analysis of the service lifecycle from the perspective of a service professional. The service lifecycle consists of:
Pre-Purchase Stage: The questions relating to the pre-purchase stage are not strictly satisfaction-related. The questions I chose pertained to how my clients found me and which of my services attracted them.
Purchase Process Stage: The purchase process questions concerned my response time to requests for quotes and information as well as the clarity of my replies.
Use Stage: During service use, clients turn their attention to punctuality and the service provider’s ability to respond to urgent requests.
Perceived Quality: Perceived quality involves the service provider’s ability to understand and meet expectations and find effective solutions, as well as the client’s opinion of the price and whether he or she would recommend the service provider to others.
With this approach, the questions to ask emerged naturally from careful consideration of each stage of the cycle.
As a consequence of this research, I began to realize that clients have a far more complex idea of “quality” than I had imagined previously. Clients consider aspects such as response time and clarity of price quotes and information, punctuality, and capacity to respond to urgent requests. They also rate the level of quality based on the service provider’s ability to understand and meet expectations and the skill demonstrated in finding effective solutions.
I had been more concerned with the accuracy and linguistic quality of translations (which can be partially equated to the ability to find effective solutions) and, to some extent, punctuality. I began to wonder if I had been neglecting certain aspects of quality that are important to my clients. At this point, finding out if they were satisfied was just as important to me as finding out what I was doing correctly.
Questions and Results
For full details of the results, please refer to the specific page on my website listed at the end of this article.2 I’ll limit myself here to a general analysis. Fifteen clients took part (38.5%), who together accounted for 73.2% of my translation income during the period covered by the survey. The questionnaire asked for 1) personal data (name of company, name of respondent, and their role or function within the company); 2) general information (non-satisfaction-related questions); and 3) my clients’ level of satisfaction with my services (the actual satisfaction survey).
1. Personal Data
Although the survey was written in Italian and addressed to Italian clients, about half the respondents gave their job titles wholly or partially in English. Two people wrote their last names in the “role or function within the company” field, which shows that some people automatically expect there to be separate fields for their first name and their family name, so much so that they do not actually read the form.
2. General Information
Questions in this section included:
How did you find Michael Farrell? When assessing the amount of work I had received and the various channels through which assignments had come, it became clear that not giving different weights to the responses according to how much work each client had provided during the three-year survey period would lead to a distorted picture of the importance of each channel. On the basis of this, I decided to calculate both weighted and non-weighted results for all of the other survey items as well. The response to this question showed just how important “word of mouth” and networking is in Italy (according to workload: 85.49%). However, due to a lack of foresight when wording the question, I was unable to establish how much work came through client-to-client contacts and how much was the result of colleague-to-colleague networking. The only other channel of any importance was having a personal website (according to workload: 12.01%).
What services have you asked Michael Farrell to provide? The fact that all of the respondents asked for translation was not surprising since it was one of the inclusion criteria. The results also showed that clients tend to underestimate the creative aspect of the work (according to workload, transcreation = 1.48%). They also don’t have a clear understanding of what localization is. As a result, I have decided to emphasize the more creative services in my range. For example, I have registered the domain name “transcreate.it” and renewed my website.
Would you recommend Michael Farrell to other companies/people looking for translation and language service providers? This is actually a satisfaction-related question (perceived quality), but since it requires a yes/no answer, rather than a score, it was tidier to include it in the general information section of the questionnaire. All respondents answered yes.
3. Level of Satisfaction
Questions in this section covered the following areas:
Response time to requests for quotes/information about the service: This was the first question with a response scale. The average was calculated by assigning a score of five points to “very satisfied,” four to “satisfied,” three to “fairly satisfied,” two to “not very satisfied,” and one to “dissatisfied.” All respondents were satisfied with the response time to requests for quotes/information, and the average was very close to “very satisfied” (according to workload: 4.83 points).
Clarity/transparency of quotes/information: All respondents were satisfied with the clarity of offers, and almost all were very satisfied (according to workload: 4.99/5 points). Providing complete and clear information is an ethical obligation. (For example, see Article 22 of the Italian Association of Translators and Interpreters Code of Practice.3)
Punctuality of translation/service delivery: All respondents were satisfied with the punctuality of the delivery of the finished job, and the average was very close to “very satisfied” (according to workload: 4.96/5 points). Besides the accuracy and linguistic quality of the translation, I have also concentrated on this aspect in the past.
Ability to respond to urgent requests: All 14 respondents who had sent urgent requests were satisfied. Eight of them (8 out of 14 = 57.14% of clients, 61.34% by workload) were very satisfied.
Ability to understand and meet expectations: All respondents were satisfied, and more than half were very satisfied (according to workload: 62.58%). It is always a good idea to find out exactly what clients expect, and this survey is an effective first step in that direction.
Ability to find effective solutions to language problems: All respondents were satisfied with the solutions I found to their language problems. I was rather upset to discover that 48.06% of my clients (according to workload) do not believe the jobs they give me pose particular problems.
Usefulness of any translator’s notes provided: All respondents were satisfied. I consider the notes to be a fundamental part of the service I provide and I am very pleased they are appreciated.
Quality/price ratio of the service: All respondents were satisfied. It is, of course, human nature to want to pay less and get the same level of service. I didn’t feel happy about asking the question about service price. It seemed fairly obvious to me that my clients would be tempted to say I was too expensive in the hope that I would drop my rates. However, after studying other questionnaires of this type, I discovered that some surveyors mitigate this risk by emphasizing the relationship between price and service quality. In other words, they do not ask if the price is fair, but if the price/quality ratio is fair.
There is no doubt the results are broadly positive. Although it was possible to leave specific comments for almost every item, only two clients wrote something, making only general positive remarks. It’s natural to wonder whether not responding to the survey in itself is a statement of dissatisfaction (56.3% of clients/less than 25% by workload). However, if natural client attrition is taken into account, together with the fact that several clients who didn’t complete the form sent new jobs after the close of the survey, this would not appear to be the case. No one chose to complain when they received the questionnaire.
I now have a totally new perspective on how my clients perceive the quality of my services. From my clients’ point of view, the concept of quality is more complex than I realized. Luckily, despite all, I don’t seem to have been doing such a bad job, but I’m now more aware of what I’m doing and—hopefully—less likely to dissatisfy my clients. The good thing is that shortly after launching the survey, work picked up again.
Are the Results Reliable?
To quote Mark Twain: “There are three kinds of lies: lies, damned lies, and statistics.”
The study population comprised 39 clients. The sample size was the number of replies I received (15). When calculating the margin of error, the confidence level, which is a measure of how sure we can be that the truth really lies within the confidence interval (the result ± margin of error), is normally taken as 95% by convention. If we feed these data into the usual statistical formulae, the margin of error for this survey turns out to be 20.11%. This is very large, but luckily the responses leaned so heavily in one direction that, in most cases, this margin is unable to transform a positive result into a negative one. As any statistician would confirm, you need a large number of responses percentage-wise to get a small margin of error when your study population (total number of clients) is relatively small.
What I Would Do Differently Next Time
I sent the survey announcement out only once, with no reminder. If I had calculated the potential margin of error before conducting the survey, I would have sent various reminders to increase participation and improve the reliability of the results. I would also define the channels through which work arrives more precisely. I’ve recently done a lot of work on defining the different types of services I offer more clearly (e.g., “simple” translation versus transcreation). I’m still not sure how to make some clients understand that translation is not always a piece of cake.
Despite its defects, which I hope others might learn from and avoid, I feel conducting the survey was an entirely positive experience from all points of view.
- Active Server Pages, https://msdn.microsoft.com/en-us/library/aa286483.aspx.
- Farrell, Michael. Client Satisfaction Survey (2011), www.transcreate.it/satisfaction-survey/
- Associazione Italiana Traduttori e Interpreti, Codice Deontologico, http://fit-europe.org/vault/deontologia-AITI.html.
Michael Farrell is primarily a freelance technical translator and transcreator, but is also an untenured lecturer in computer tools for translation and interpreting at the Libera Università di Lingue e Comunicazione (Milan, Italy). An Atril-certified training partner, he is the author of A Tinkerer’s Guide to Structured Query Language in Déjà Vu X. He developed the terminology search tool IntelliWebSearch. He is also a qualified member of the Italian Association of Translators and Interpreters. Contact: firstname.lastname@example.org.