If a brand overhaul or refresh is in the cards for your company, there is a lot to consider. At the top of your list should be the perceptions, preferences, and opinions of your customers and internal stakeholders. And the most effective way to gather this information is through an anonymous survey.
In Part 1 of this two-part blog series, we’ll look at the basics of survey construction. In Part 2, we’ll take a more in-depth look at psychological factors that can help you develop better questions.
A well-constructed survey can be invaluable to branding, marketing, product de¬velopment, and HR projects. A poorly constructed survey, however, can be a giant waste of time for both you and the people you ask to take it.
When it comes to customers or prospects, it’s even more important to make the survey easy and painless to complete. Good preparation and question construction are key to a survey’s success. Here are some tips for constructing a thoughtful survey.
Find your purpose
Narrow in on the answers you are looking for by asking yourself exactly what you need to know, in a very specific way. Vague questions will get vague answers. For example, if you ask “What do you think of my company, you will likely get answers like “They’re great” or “Good products”. But if you’re wondering why people aren’t buying more of the things you’re selling, or you suspect you have a customer service issue, you’ll want to build questions that elicit more specific answers (without being leading).
Define your goals
Think about your goals in terms of your future actions. What do you want to do once you’ve analyzed your survey data? What strategies are dependent on knowing the answers to the questions you will be asking?
Indentify your audience
Determine who you need to survey to get the information you need. Customers, employees, website visitors? Think about the number of participants you’ll need to get a good response. Not everyone will complete the survey, so find the largest pool of people who could be expected to answer your questions in a meaningful way. Participation is the number one goal of survey construction. The more participants you have, the more relevant and significant the results will be.
Decide on delivery
SurveyMonkey and other online tools are great resources for administering and analyzing surveys online. Simply set up your survey questions and answers and provide your participants with a link via email or your website. In-person or phone interviews are also an option, though they are much more time consuming to set up and administer.
Draft your questions
Keep your questions clear and concise. Ask one question at a time and don’t assume your audience has all the knowledge you do. If you need to define terms or simplify language, do so.
When your first draft is complete, consider whether each question should be structured as open-ended (free-form answer) or closed-ended (true-false, yes-no, or rating/ranking systems). Closed-ended questions will give you easily quantifiable data. Open-ended questions will give you more detail, but will require more analysis and aggregation. Generally, if you can provide simple two-choice answers or a rating or agreement scale, your answer could be structured as closed-ended. See Part 2 for more details about question construction.
Refine and edit down your questions
When you’ve finished assigning response options to your questions, review your list for length, depth, and time requirements. There is a difference between asking people to answer a handful of true-false questions and asking them to explain their theory on a particular issue. If your questions are complex enough to require a series of open-ended responses, only ask the questions that are vital to your goals.
If you think it will take more than 10 minutes for someone to complete, consider trimming some questions or offering a participation incentive. Abandonment rates soar with longer surveys.
Beta test your questions
Evaluating your questions for clarity and reliability is an important step. Ask a few colleagues to take the survey while you watch. If they ask you to explain the meaning of a question or the kind of answer you’re looking for, or if it takes more time to complete than you expected, you’ll need to make some revisions.
Once revised, it’s a good idea to test again with someone who hasn’t seen the questions in their earlier form. This will highlight any final tweaks you need to make to clarify questions or answers or adjust scale mechanisms that are too complex, too simple, or inconsistent from question to question.
Proof and deploy
A simple typo can change the meaning of a question or answer and invalidate responses, so proofread everything before deployment. You should also consider setting a deadline for completion and sending a few reminders as the deadline approaches. Then, push send and sit back and wait for all that useful data to come rolling in.