While it is far more preferable to do a customer visit interview, customer surveys offer an opportunity to poll your users and potential customers on questions that might otherwise go unanswered. But to be effective, it is necessary for you to have a plan as to how to get the best out of your surveys. The following are some of the suggestions I have found to be effective at developing robust and meaningful customer surveys.
Think about your target audience—Who are they and where/when/how will they respond to your survey? Is your ideal survey respondent an executive concerned about the bottom line of his/her manufacturing plant, or is it someone who is interested in making his or her life easier or more enjoyable? Will they respond to your survey on a PC or on a phone? Grid questions work well on a PC, and you are better able to present a slightly longer survey. Mobile phones, on the other hand, do not support grid questions very well, and the expectation is for a much shorter survey.
Choose your preferred survey method, and how best to distribute your surveys—As discussed previously, a computer-based survey may be better for surveys requiring a more detailed response, but what about respondents who are not digital natives or computer savvy? In that case, a paper, mail-in survey may be preferable. If you are targeting existing users, an online survey on your website may suffice, but if you need to reach new consumers, a paid ad targeting specific customer demographics and allowing you to target specific people may be a better option:
- KISS (keep it simple and short): Find the shortest way to ask the question without losing its intent. There is a direct correlation between survey length and rate of abandonment. Research has shown you can expect a steep drop in responses if your survey takes more than 5 to 10 minutes—if you want people to finish your survey, be respectful of their time.
- Have a plan for every question you ask: Every question you ask must have a well-defined purpose and a strong reason for being asked. Make sure you also fully understand how a customer's response will drive changes in your research. If it doesn't matter how they answer it, why are you asking it?
- Get buy-in from other parts of the organization: Discuss the survey with other departments, such as marketing, product management, sales, engineering, and so on. Managers from other departments may have insights you haven't considered. Nothing is more painful than doing customer survey research when you are missing that one vital piece of information because you neglected to ask a specific question.
- Construct smart, open-ended questions: While it is easy to ask yes/no questions or multiple choice questions, by far the most enlightening responses will come from survey questions that allow customers to describe, in their own words, their responses. However, the survey structure must also be a consideration as nothing is quite so intimidating as receiving a survey with a large text block connected to the first question.
- Make use of yes/no questions: While the open-ended text questions will give you richer data, most of the time a yes/no response is all you need. It has also been shown that closed-ended questions make for great starter questions because they are typically easier to evaluate and complete. A good rule of thumb is to put demographic questions first and make them mandatory. Then, transition from general to more specific questions. Open-ended questions work best at the end of the survey, allowing respondents to provide comments or express opinions.
- Ask one question at a time: Do not group questions into the same query. Don't ask questions such as How did you find our product? What did you like best and least about it? Why or why not? Multiple point questions greatly limit your probability of getting answers to any of the questions.
- Make rating scales consistent: If you ask one question and the directions are for the respondent to choose between 1–5, with 1=Strongly Agree and 5=Strongly Disagree, don't ask another question later in the survey with scales in the opposite orientation like 1=Least Important and 5=Most Important.
- Avoid leading questions: Questions that point respondents to answer a certain way may make you feel better about your product, but won't give you the data you need. You should avoid phrasing the question as We have recently upgraded our capabilities to be the most powerful widget you can buy in the world. What are your thoughts on our new features?. Instead, you should ask What do you think of the new widget features?.
- Avoid acronyms and jargon: Unless your surveys are extremely targeted and customers are very knowledgeable about your industry, you are likely to run into problems as the customer may answer untruthfully or incorrectly if he/she does not understand what you are asking.
- Consider reporting back to the customer: Quite often, a customer will be interested in understanding their perceptions versus those of their peers. By offering the chance to see how they answered the questions compared with other respondents, you will increase the participant's engagement level and thoughtfulness of response.
- Test your survey: Once you have gone through the process of creating your questions and planning your survey, conduct one last check to make sure everything is working as it should by doing a pretest with your coworkers or a few members of your target audience to make sure there are no glitches or unexpected question interpretations.
Once you have collected all the data, it will be necessary to analyze the results. This is where online surveys really shine as most have built-in analysis capabilities, including tables, charts, graphs, cross-tabulations, and more advanced functionality. With most of the tools available on the market, you can also export it into Excel to build bar charts or pivot tables that help you draw more meaningful insights.