Creating a Customer Survey

Customer feedback offers a wealth of information that can help you steer your business toward greater success. By understanding your customers' pain points and relieving or eliminating those, you provide your customer with more incentive to stay with you. You may also need to understand the priorities of their needs. You can use simple survey tools to get this information. Before we talk about the tools you can use, let's talk about designing a simple survey.

• Determine the business questions you want to answer

If you have an e-commerce website, tracking statistics can give you information about where your site visitors come from, what they look at on your website, and what percentage of visitors conclude their visit with a purchase. If you have a service business or otherwise do not sell your products directly online, you may want more information about the decision points your customers make (or may not make) toward purchasing your service or product.

For example, you may want to know how potential customers find you. You may want to understand how customers choose to contact you versus your competitors. You may want to understand why customers choose your product or service instead of your competitor's.

Or you may want to find out about how customers respond to your service representatives, or if they continue to be happy with your product or service after the sale.

Make a list of the questions you want to answer. Understand what actions the answers will help you take to improve your business.

• Create a distribution list for your survey

If you have actively collected names for a mailing list, then you have a good start for a distribution list for your survey. You'll want to find the correct segment within that list who can answer your questions.

For example, if you are trying to understand why customer chose your products, you'll want to send the survey to individuals who have actually made a purchase with you. If you are trying to understand how to attract new customers, you may want a combination of existing and potential customers. In this case a potential customer may have registered at your site but may not have made a purchase yet.

• Set expectations for your survey subjects

To increase your response rate to your survey make sure you set clear expectations upfront. For example, start your survey by asking "To help us improve our service, can you take about 15 minutes to answer these 10 questions?" Then, of course, follow through and make sure the survey only takes 15 minutes to complete and indeed only has ten questions.

• Create questions that will give you clear, definitive answers

There is a whole science of questionnaire design. Let's talk about a few simple things you can do to give you more meaningful results for your survey.

Avoid leading questions on your survey.
Example of leading question: Some of our customers find our return policy difficult to understand. Did you?
Example of non-leading question: Please rate the clarity of our return policy: very clear, somewhat clear, not very clear, not clear at all, have not read the return policy

Understand the real goal of your question. For example, you may want to understand if your customer likes your product or service. This is a reasonable question. But, how much do they like it? Instead of only getting their experience with your product, consider whether they might become your advocate for an additional sale. In that case, you could ask questions like "Would you purchase this again?" or "How likely are you to recommend this product to your friends?"

• Test the survey

Now that you have narrowed down your ten questions, test your survey on a couple of people who will give you honest, helpful feedback. This step will help you avoid confusing questions and can also help you time how long it takes to complete.

• Send the survey out and set a response deadline

Now you are ready to send out your survey. Depending on your sample size, there are a few free options you can try.

The Basic Plan from SurveyMonkey allows you to create a 10-question survey for up to 100 respondents for free. You can do this an unlimited number of times.

Zoomerang also give you a free plan for 12-question surveys and up to 100 respondents. Also an unlimited number of surveys.

SurveyGizmo offers a free service with a slightly different structure. You can have an unlimited number of surveys with an unlimited number of questions, but the limit comes into play for the total number of respondents across all surveys per month at 250.

There are other free and low cost survey tools that may fit your specific needs. Be sure to understand your survey tool requirements before you pay for a service.

• Interpret the results

You've sent out the survey and responses are starting to come in. Time to look at the results.

How is your response rate. Did you provide a good subject line that enticed your customers to open the email? Did you send the email out during the middle of the week? (Survey response rates are highest when your target respondent is actively on their computer but not too busy catching up or trying to finish other things. Typically this means Tuesday, Wednesday, or Thursday first thing in the morning or at noontime are good times to send out a survey.) Did you set the expectations at the beginning of the survey for length and time commitment?

What actions are your respondents encouraging you to take? More of the same and stay the course? Or, rethink your strategy in sales, marketing, or service? If you've designed your questions well, your next steps should be clear.

Last Notes
If you want to know more about questionnaires and surveys, there are books devoted to the topic including how to avoid bias, how to effectively use different question formats, how to avoid bias in your sample demographic choices, how to elicit the most information before revealing information to the respondent, etc.

In the meantime, you have enough to get you started. Every bit of information you can gather, whether formal or informal, helps you improve your business.