Understanding your customers’ sentiments is the first step to customer service improvement. However, it would be difficult and impractical to talk to them one by one to accomplish this. So, unless you can develop the power to read minds, the next best thing would be to utilize a customer satisfaction survey.
Many times, surveys are conducted manually using printed forms that are handed out after each transaction. Alternatively, you can digitize and automate the process using customer support software. However, before you leverage surveys in your operations, it is best to understand that there are various survey-taking guidelines and best practices that you should consider.
In this article, we’ll guide you on how to create a customer satisfaction survey from writing questions to providing answer choices. We’ve compiled tips on yielding more objective results and gaining additional insights from surveys. This way, you can learn how to measure customer happiness accurately and identify pain points across all your touchpoints.
Customer satisfaction has become more critical today than ever. With the prevalence of social media and the group-think phenomena, many companies feel that they are at the mercy of their customers. After all, nowadays, one viral tweet can easily make or break a business.
To wit, there is really no way for a business to know and understand how it impacts customers’ lives without asking questions. This is why market research and survey results are prized commodities in business communities. And, thanks to technology, they are easier to conduct and analyze now. For instance, we know that email surveys optimized for mobile generate 15% more clickthrough rates. It is close to impossible to get a diagnostic like this from traditional pen-and-paper survey forms.
However, there are pros and cons to surveys. The tool is useful, but it is not airtight. Moreover, it does not always give you a complete picture of the situation. Surveys may also be cost-efficient and scalable. However, it can be prone to dishonesty, biases, and hidden agenda.
Still, the pros still outweigh the cons if you do the survey right. Thus, in the next few sections, we will discuss the basics of surveys and how top-performing businesses make use of them.
Surveys are used in many ways and forms by different institutions, private or public. However, there are mainly two kinds of surveys used in human research: (1) informal and (2) formal.
The former doesn’t make use of questionnaires. It is mainly informal interviews from people. What’s great about it is that it is dynamic and iterative. Also, it gets better qualitative data from real-life conversations. On the other hand, formal surveys use questionnaires and get better quantitative data. It has lower sampling errors, and it permits statistical analysis.
Moreover, formal surveys require hypotheses beforehand. It is not like informal surveys where there is much potential for learning. Formal surveys are designed mainly for verification. You already know what you are looking for. You just want to know whether your assumptions are right or wrong to some degree. In the case of customer satisfaction surveys, you assume that you either are doing well or not.
As you can see, both types of surveys have advantages and disadvantages. In the following subsections, we’ll discuss common surveys that marketers and customer support specialists regularly use.
Whatever survey type you choose, technology can help. Most help desk software solutions today allow you to ask questions to your customers, both casually and formally. Casual questions are as simple as inserting your intended questions during customer conversations. So what you need to pay attention to is the app’s formal survey feature.
To give you an example Freshdesk lets you automate CSAT surveys (more about this survey type below) across all digital touchpoints from social media to email. You can also have an option to send multilingual CSAT surveys. This is perfect for multinational brands or those that cater to a diverse mix of customers.
You can check the features closely when you sign up for Freshdesk free trial here. This is a good way to benchmark how technology can help boost your customer survey accuracy and efficiency.
CSAT is a classic survey that has stood the test of time. Every CSAT survey should be different for every touchpoint or experience. Well, this is the ideal, and it is a full-time job to actually get this optimal. However, all CSAT surveys have the same purpose. They want to measure the satisfaction of customers for a given service, product, or touchpoint.
Thus, analysts and marketers don’t need to reinvent how to create a customer satisfaction survey fully. All they have to do is take a cue from successful ones done before and make it their own.
Usually, CSAT surveys are sent through different channels. They can be deployed offline and online. Many firms often send out one survey question using a five-point, ten-point, or seven-point scale that goes something like this: “How satisfied are you with your customer service today?”.
However, when we keep the best practices and tips in mind above, this is too broad. Thus, it is recommended that you add qualifying customer satisfaction survey questions.
Perhaps, you can add questions about how they rate your digital interface, how often do they read your content, and how they rate the quality of your content. This way, you get more actionable data. Just remember, keep the survey short.
CSAT surveys are usually conducted within a customer service or support context. Customers are usually given survey questions after their interactions with service or support representatives. These questions are usually given through email, webchat, or social media today. Many still do this manually and record them on spreadsheets. However, there are tools out there for customer service teams to use.
There are standalone survey-taking programs. But, there are tools specific to the customer support and service context with CSAT survey capabilities, as our example above, Freshdesk. What we mean by these are full-blown help desk solutions that can also automate survey-taking processes like sending questions, tallying answers, and generating reports.
Moreover, other top-performing businesses also include what’s called the Customer Effort Score (CES) survey in their CSATs as well. They use this to gauge customers’ ease of experience. Why? Surveys only measure customer satisfaction with the support given to resolve some issues. This doesn’t zero in on how easy or accessible it was for the customer to get the issue resolved. Thus, it is better to ask a CES question as well.
One way to ask the CES question is: “How do you rate your effort to solve your issue today?” Then, provide the seven-point scale from very easy to very hard. Remember, do not use the word “easy” or “hard” in your question. This might prime your customers to answer in a not objective fashion. We’ll discuss how this works later.
The CES and CSAT bundled survey allows you to get more information. The former allows you to gauge how satisfied they are with the resolution, your agent, and the channel. The latter gauges the effort the customers have to put in to resolve their issue. In this way, you get insights into more customer pain points and act on them.
Customer satisfaction does not necessarily equate to customer loyalty. Remember, most CSAT answers are event-specific. They only measure the most recent interactions. However, there is one way to check for brand loyalty indirectly. It’s through the Net Promoter Score (NPS) that directly measures brand advocacy. Why? Loyalty customers are more likely to advocate for a brand to their families, friends, and colleagues.
Clever marketers use NPS to identify loyal brand advocates and measure the percentage of their customer base that is likely to recommend their brands. The NPS score is measured using a ten-point rating scale where one represents “very unlikely” and ten “very likely.”
Those who score ‘nine or ten’ are Promoters. They are the most passionate and loyal customers that you have. Those who answered ‘seven or eight’ are considered to be Passive. They are not really too excited about the brand enough to promote it. Lastly, the respondents who answered ‘zero to six’ are considered to be Detractors. These are the customers that are least likely to promote the brand.
The usual NPS question goes like “Considering your total experience with company X, how likely are you going to recommend company X to your colleagues, friends, and family?”
Traditionally, this is a singular question. However, you can also add two or three more. You can make them specific”, like how they are likely to recommend the brand on Twitter, Facebook, Instagram, etc.
Informal surveys are usually structured interviews. Researchers use this to derive information with better depth and breadth. Why? In these face-to-face interactions, customers respond to open-ended questions more freely. Thus, they provide a richer account of their brand experiences.
We highly recommend that marketers conduct informal surveys from time to time. This can be a bit costly compared to digital surveys. However, they provide stories or narratives that can help explain the numerical data from formal surveys. With this, marketers get more chances to observe customer interactions and pain points firsthand. Also, this allows them to draw detailed answers from customers. Thus, they get to know the reasons for customer dissatisfaction and other issues more intimately.
Lastly, high-performing businesses always conduct informal surveys together with their formal surveys. This, we believe, contributes a lot to their success.
There is an art and a science to survey creation. As such, over the years, many experts have established some rules of thumb as to how it should be done. These are guidelines that we all should keep in mind before we create surveys and the questions therein. Here are some of them:
These rules of thumb enable you not only to get useful answers from your customers. They also allow you to get your customers to entertain your surveys. After all, there is no point in making surveys that people won’t answer.
Survey questions are really what makes up formal surveys. These allow you to prod customers’ perceptions and feelings about your brand and your touchpoints. Thus, these need to be clear and concise. They also need to be specific and measurable. Moreover, survey-takers use two types of questions: closed-ended and open-ended questions.
Closed-ended can be answered categorically. Common forms include the multiple-choice form and the rating scale. On the other hand, open-ended questions allow respondents to answer the questions on their own terms. However, they may take a longer time to answer and, more often, a longer time to analyze. Thus, if a marketer follows the rule of thumb that the customer’s time is valuable, he or she will opt for the former.
However, expert survey creators keep open-ended questions optional. In practice, they are located at the bottom of the survey form with an optional field. So, with the desiderata in our best practices box in mind, let’s breakdown how to write effective survey questions and answers.
A bad survey question is as follows: “Do you use product X a lot?” Combine this with only two answer choices: yes or no. Then, you get a broad answer that can be interpreted in different ways by different people. Some people may define “a lot” differently than others. Others may find “a lot” to mean more than once a day. Some don’t.
Also, remember that you don’t only write for respondents. You also write for internal interpreters of data. These include yourself, your team, upper management, and adjunct teams. Thus, we need to do better than this. Let’s keep the following in mind.
Considering these, we can now create a better version of the one before. One of which comes in the form of “How often do you use product X?”. Through this, we can verify that it measures one underlying concept, which is the frequency of use of a certain product. Also, no other concepts are measured. So we can safely say that this question means the same to all (sane and honest) respondents.
Pair this with a good set of answer choices like “(a) more than once a day, (b) at least once a day, (c) two to three times a week, (d) once a week, (e) other: _____.” Using this set of options, you can get a more specific answer. Thus, your firm will get a more specific insight on how it is used.
When providing answers to close-ended multiple-choice questions, the first tip is that your set of answers should anticipate all possibilities. If you noticed in the previous section, the example provided for an answer set anticipates different frequencies of use. However, this does not cover every plausible answer.
Thus, we added the “other-specify” option. In this way, you also cater to more possibilities and even open-ended answers. Don’t simply make “none of the above” one of the categorical choices. You won’t get anything from it.
Moreover, like the question above, many questions can be answered using scales. This is especially so when we want to ask attitudinal questions like, “Are you satisfied with the service you received today?”. Maybe, a good yes or no answer will suffice. However, if you want to be more precise, you can use rating scales.
Here, experts suggest using either a five-point or seven-point rating scale. The least numeral represents the lowest rating, and the largest one represents the highest. Abstraction is key in this one. The reason for this is that humans have certain cognitive quirks and biases.
One example of this is when we are asked questions in the form of: “Do you agree with X?” or “How easily did you access our support today?” The words ‘agree’ and ‘easily’ prime us already. In this way, they are leading and can be written more objectively.
Thus, it is best to avoid such words. One can go about it by writing questions that both have positive and negative items that can be scored “high” or “low” on a scale. This level of abstraction mitigates biased answers and helps us get more objective answers from customers.
We know that questions are like the building blocks of survey questionnaires. However, how you string these questions matters too. The same goes for answers. This is because of what some scientists call cognitive biases that sometimes get a bad rep. They do have their uses, though. However, this is not the place to discuss them. For our purposes, it is enough to know that these biases can affect the objectivity or your survey results.
Biases such as the priming and recency effects can be induced by poorly-designed survey questionnaires. We’ll discuss how these biases can ruin your surveys in the next subsections.
One is that you have to keep your question order in mind. It’s because survey answers can be impacted by previous questions. This is caused by a psychological phenomenon called priming. This is when exposure to a stimulus affects how a person responds to another related stimulus.
For example, question one asks: “In the last six months, have you experienced service downtime?”. Then, the second question goes: “Using the rating scale below, are you satisfied with your current plan?”.
The first questions already “leads” the person to remember bad experiences in the past six months. Thus, this may likely affect how they answer if they are satisfied with their current service. They might have had answered differently when the order of questions is interchanged.
So, the issue is not just about writing questions themselves. The order of your questions also matters.
In psychology, researchers explore the effects of the primacy and recency of stimuli. Both are related to how we contextualize the information we get. However, these biases also skew our judgments towards certain options just by the order they are presented. Put roughly, some involuntary cognitive mechanisms sway us to answer one way over others.
The primacy effect is where a person tends to remember and choose the first option over the latter ones. Neuromarketers leverage this to increase sales. Yes, there is a field of that name. Also, according to the Harvard Questionnaire Tips Sheet, the primacy effect rears its ugly heads in both paper and internet surveys as well.
On the other hand, the recency effect is when subjects tend to choose the most recent stimuli more readily than previously shown options. This mainly affects internet surveys that use drop-down menus to answer choices.
Thus, experts suggest two ways to mitigate these problems:
We’ve discussed some tips on how you can design your questions, answers, and their orders. Now, let’s get into other parts of conducting a survey: measuring and interpreting data.
In practice, this has been simpler thanks to online tools such as digital survey solutions. Also, many use business intelligence platforms to get advanced insights.
However, interpreting survey data with other company information is still a different skill to master altogether. To help you learn how to do this better, we’ve provided examples of techniques to generate actionable insights from your surveys below.
Normally, customer satisfaction surveys are given periodically. Ideally, you should give them every day and get responses for every consumer interaction with the brand. Then, you’d get this data then compare it with other sets of seemingly exclusive data to get better insights.
This is called cross-tabulation (you can do it somewhat easily using Excel). With this technique, you can find combinations that may be useful for your operations.
For example, you have given out CSATs every day for six months. You notice an uptrend in customer satisfaction, which is good. However, you also notice that it oscillates at regular intervals. So, you compare this or cross-tabulate this with other data like staff rotation. Then, you found out that whenever employee Y is at work, the CSAT goes down. Also, when you add another set of data like the number of CSAT surveys sent via email, you see the number go down as well. This data is somewhat consistent for the past six months.
What could this mean? Well, one interpretation is that there might be something wrong with what is employee Y doing. So, you investigate the issue further to see if employee Y is indeed the problem.
However, after investigating further, you found out that employee Y is also a part of a small task force for product development. This could also mean that he is getting spread a little thin. So, is he the problem, or do you lack personnel?
This is a hypothetical scenario. But, hopefully, this gives you an idea of how to use CSAT results with other data to gain actionable insights. You don’t only measure customer satisfaction, but you also find out what could have been the causes of the scores by looking at correlations. Then, you can actually do something about it.
Another simple way to get insights from your CSAT surveys is by using a line graph or a bar graph. You can do this on Excel easily. For example, you can use a line graph to diagnose which parts of your customer journey provide under par customer support.
What you do is line up your CSAT scores (in a seven-point scale) onto time intervals like weeks or days. So, you’d see in which weeks customer A graded you well or poorly. Then, you label these intervals according to your customer journey map. Preferably, this map is made up of several phases, from being a prospect to a loyal customer. Now, you’d see how satisfied customer A is with your service per phase.
Do this for more customers, and maybe you’ll see a general pattern of where your service has weak points and strong points in your customer journey. Moreover, you can cross-tabulate these with other data sets and find more actionable insights, as discussed in the section before.
For a bar graph, you can use this to see the total customer satisfaction tallies for a single customer or all of your respondents, as you can see in the picture below. This helps you have a general picture of the total health of your support or marketing efforts.
High-performing businesses use surveys for various reasons. But remember, they do this to gain insights from customers on how the firm can provide better value for them. So, in a strong sense, these revolve around customer happiness and satisfaction. The following are common reasons why businesses perform surveys:
Often, high-performance companies create surveys with different objectives in mind. Many try to hit a few birds with one stone. For example, they create a survey quiz to increase engagement while also gathering new leads.
However, remember that surveys, first and foremost, help the company understand customers’ needs, wants, and expectations and how to satisfy them. Thus, we always design surveys with this goal in mind. Also, we need to deploy surveys for and in various business touchpoints as well. Meaning, you need your customer satisfaction survey template to be flexible.
High-performing businesses understand that they have different touchpoints. They also know that customer experiences on these touchpoints vary. So, they try to understand the interactions per touchpoint to make marketing and sales strategies more targeted. That said, it is important to conduct surveys on different channels to gain higher quality data and insights. Some of the avenues that you should consider are:
In this way, businesses can measure the performance of each channel and identify pain points for customers. As a result, they can gain actionable insights and optimize their operations.
Thanks to digital web-based tools, marketers can easily create, send, and measure survey questionnaires. Research companies or companies with big marketing departments usually use dedicated tools for survey management. However, if you don’t have the budget for these, you can always use quick and easy tools such as:
You can use these to easily gauge customer sentiments and satisfaction about products or an event. After all, they are native to these channels. So, you’d hit your target audience for each channel, as well. If you are a starting entrepreneur, you can also take advantage of top survey platforms that cater to small businesses.
However, if you are a company that regularly gets inquiries or support issues, you may need a full-blown help desk software like Freshdesk as mentioned. In this way, you’d automate sending CSAT surveys, generate reports, and get insights regularly in real-time. However, tools are just tools. They work only as well as the people who use them. Thus, when customizing surveys, keep in mind the best practices and tips above.
It is high time to improve customer-facing processes. This is especially true as US customers are less satisfied than their foreign counterparts. Yes, customer satisfaction is better overseas. If you need to acquaint yourself with facts and figures, please read our report on customer support satisfaction.
After all, we don’t want to lag in this regard. Thus, we need to ask our customers what they want and how they feel about our products and services more and more often. This is where you’ll need to implement customer satisfaction survey campaigns.
However, before you do so, you must remember that surveys are just tools to help you know how you can provide more value to your customers (at a profit). You should keep in mind that customer success is your top priority here. Therefore, when you plan your survey efforts, make sure that everything in there is customer-centric. What this means is you try to drill down into customer pain points and ask questions on how to extinguish them. This goes for both online and offline surveys. To wit, we’ve included general steps and tips on how to prepare a successful survey campaign.
If you don’t have the time to prepare this yourself, you can take advantage of popular digital survey solutions. Templates and workflows are natively available on these platforms. Alternatively, if you are a customer support specialist that wants to get CSAT, CSE, or NPS survey results regularly, you can do well with a top help desk solution.
FinancesOnline is available for free for all business professionals interested in an efficient way to find top-notch SaaS solutions. We are able to keep our service free of charge thanks to cooperation with some of the vendors, who are willing to pay us for traffic and sales opportunities provided by our website. Please note, that FinancesOnline lists all vendors, we’re not limited only to the ones that pay us, and all software providers have an equal opportunity to get featured in our rankings and comparisons, win awards, gather user reviews, all in our effort to give you reliable advice that will enable you to make well-informed purchase decisions.