Customer Service Survey Questions You Should Ask to Get Good Data

9 min read
Jun 6, 2016
  • Post on Twitter
  • Share on Facebook
  • Post on LinkedIn
  • Post on Reddit
  • Copy link to clipboard
    Link copied to clipboard
Customer Service Survey Questions

Customer service surveys allow you to take a peek into your customers’ minds. You can learn about their problems, needs and do something about them. It’s your very own customer service magic. But instead of a crystal ball, you need a few, well worded customer service survey questions.

Without some form of a survey for your customers, you are forced to guess and make estimates when it comes to your customers’ satisfaction. And going only by your gut when making business decisions is not the most reliable option.

Start making more informed business decisions by collecting customer feedback. See what customer service survey questions you should ask and how you should do it to get the best results.

Customer service survey questions ideas

To start off, we first need to discuss the most popular survey questions and why we really ask them. These basic customer service survey questions will give you an idea what to ask your customers about. Once you get familiar with these, we will move onto the finer details of creating surveys.

Keep in mind that these are more like directions rather than templates. You can word them however you want, but the general aim of these questions should stay the same.

Asking for personal details

Usually the first questions you ask your customers is their name, email address, phone number or anything else that can help you identify them. However, before asking for all these, you should first think if you really need all this information.

You will need some form of identification. There’s no question about it. Keep in mind that, depending on the circumstances, some of this information can be already available to you. For example, if you offer a customer service survey via email, you obviously don’t need to ask for the customer’s email address.

You shouldn’t offer the same customer service survey to all customers. Even if it is easier for you, you should change the survey for different channels. Asking for email makes sense when you offer the survey over chat but not when it is sent via email.

If you offer user accounts on your website, you can probably do away with most ‘identification’ questions and offer surveys to customers who are logged in (unless you want to get some additional personal information, like asking for phone if you already have the customer’s email).

Customer satisfaction questions

The most basic way (that works pretty well) to gauge your customers’ satisfaction is to offer a binary thumbs up and thumbs down rating option. This input will allow you to calculate customer satisfaction with this formula:

Customer Service Survey Questions Formula for Customer Satisfaction

If you need more details, you can offer a broader rating option. For example, you can use a range that goes from very bad, through bad, neutral, good to very good. A range like this will also give an option to customers who feel that the service was neither good nor bad.

There’s also the possibility of simply asking the customers for opinion in an open-ended question but that can be a lot more difficult to analyze than a simple good/bad answer.

Agent efficiency survey questions

Another question you could ask when it comes to customer service is about the case itself. Asking whether the case was resolved during the customer service interaction will allow you to see how successful your agents are. It’s entirely possible to have good customer service experience and not have one’s case resolved. And the other way around too – you can resolve a case in a way that could be handled better. Getting this extra bit of information will show you which cases are the toughest and which interactions need improvement.

Apart from asking about the case resolution, you can also ask if this is the first time a customer contacts you about a particular matter. This allows you to see how many of your interactions end in one touch and how many need a follow up contact.

NPS customer service survey questions

NPS questions are one of the more useful ones you can put in your customer service surveys. These are only two additional questions that can provide you with very useful information: the scope of your customers loyalty. The two questions are:

  1. How likely are you to recommend our product to your friends and colleagues?
  2. What is the reason for your rating?

In the first question, you offer a 0-10 scale. Depending the rate your customer picked, you assign them to three different groups.

The customers who gave you a 9 or a 10 are promoters. They are not only satisfied with your business, but they are satisfied enough to spread the word and recommend it to others.

The customers who gave you a 7 or an 8 are passives. They are happy with your business but you still need to amaze them with your service to turn them into promoters.

Customers who gave you 0-6 are detractors. They are either plain unhappy with your service or are not happy enough and will churn out over time.

The second question provides you with a reason why somebody gave you the score they did. In the case of promoters, this is useful because you can discover what experience turned them into a brand advocate. Once you learn that, you can try to replicate that with other customers to turn more of them into advocates. In the case of customers who gave you a poor score, you can learn what you should improve to keep them from leaving your business.

NPS questions can give you a lot more information than a plain old customer satisfaction question. They put your relationship with a customer in a perspective. For example, a customer who gave you a 7 or even a 6 would probably say that they are satisfied and give you a thumbs up. However, there’s a huge difference between a thumbs up from a promoter and a thumbs up from a passive or a detractor.

Open-ended vs. close-ended customer service survey questions

The type of the answer a customer can give in a survey should depend on the goal you want to reach with the survey. You have two basic options here.

The open-ended questions are definitely tougher to swallow for customers. They require customers to stop and think for a moment before typing in their answer. Not a lot of customers will give you that much time when providing their input. Unless they are really loyal, most customers will skip tougher open-ended questions. The upside of these questions is that they can produce results you would never expect.

On the other hand, you have the close-ended questions. As you can imagine, they won’t produce the same amazing, detailed responses. Since you preset all the responses yourself, customers won’t be able to surprise you. The benefit of using close-ended questions lies in the sheer volume of responses you can get. The time it takes to answer one open-ended question could be spent to answer ten close-ended ones.

One thing that could sway your decision on the type of answers you should offer is the way you want to process data later on. If you’re planning to use it for some KPIs or other metrics, you’re pretty much limited to close-ended questions as you can easily convert them into numerical values and analyze in a spreadsheet. Open ended questions are perfect if you want to gather some in-depth feedback or discover new ways of helping your customers.

Look for opinions that make sense

In your quest to get as much survey data as possible, you may be tempted to send a survey to literally everyone who visits your website. Before doing that, you need to ask yourself: do you really care about the opinion of everyone?

Wouldn’t it make more sense to only ask the people who, for example, already bought something from you? By surveying your customers, you get much more valuable data than by asking every random visitor who reached your website.

Your customers bought something from you for a specific reason. You should try to discover that reason to recreate the same experience that, hopefully, will lead to a sale too. Asking every random person who stumbles upon your website to fill out a survey will tell you everything BUT that one important reason.

Keep in mind who you are trying to survey and what kind of information you can get from those people. Asking existing customers makes a lot os sense but asking people that just got to your website about their experience is a waste of everyone’s time.

Keep the goal of the survey in mind

You should use several different surveys for different goals. If you try to fit every question you have into one survey, you may end up with very few results as it will be too big of a bother for your customers. What you can do instead is to segment your surveys a bit and keep the questions related to each other.

For example, if you want to gauge a customer’s satisfaction, you shouldn’t throw in a random question about the way somebody got to your website.

Each survey should have a clear goal. Your customers should be able to tell why you need the data and every off-topic question will make them feel more and more unsure. To increase your chances of getting a survey filled out, you should clearly state why you need the data and how do you plan to use it.

For example, you can let your customer know from the get-go that you simply want to find ways on how you can improve their experience when dealing with your business.

How to get people to fill out your surveys

Apart from letting your customers know why you need the data, here’s a couple of additional ways you can get more results:

  1. Branding your survey: Something as simple as using your company logo and colors in the survey’s design will help your customer recognize your business and make them more likely to fill out a survey.
  2. Offer incentive: To get more surveys filled out, you can also offer an incentive of some kind. For example, you can offer a discount on customer’s next purchase after they fill out your survey. In such cases, you want to be able to identify the person in some way. If you have no way of verifying these results, people may provide some random answers just to get the discount code.

The most important aspect of using customer service surveys

What’s the single most important aspect of running a customer service survey? You should end up using the data for something.

Each time you ask a customer to fill out a survey and the customer complies, they give you a favor. It may be a small favor, but it’s one nonetheless. After all, customers go out of their way to spend some time on your survey. It would be a shame to have all those favor and all that time wasted if you end up not using the data to improve your business.

If you want to set up a survey, make sure you really need the data and that it makes sense to ask your customers for it.

Get a glimpse into the future of business communication with digital natives.

Get the FREE report