Tips on How to Build a Great Survey
If you’re interested in using QR Codes or Personalize URLs then you’re probably considering including a survey on the landing page. If that’s the case here are some good tips on how to build a quality survey that will deliver the results you want.
1. Define your objective
What business goal do you hope to achieve with the survey data? This objective should be clearly stated and should guide every survey question you write, how much detail each question needs and if there’s any way to access the information elsewhere to avoid unnecessary questions. Without a clear objective, surveys become too long, complex and expensive.
2. Choose the right contact method
Traditional survey methods rely on face-to-face interviews, phone calls and mail questionnaires. But e-mail and online surveys are the fastest-growing and often the least expensive to administer. Choose the method that fits well with both the nature of your questions and the depth of information you hope to get. Open-ended questions work best in-person or over the phone, while questions that require specific rankings, true or false answers or particularly candid responses are better via e-mail or traditional mail. Remember, when deploying a survey through e-mail, make sure you have valid opt-in permissions to avoid complaints about spamming.
3. Write the questions carefully
Developing an effective and easy-to-answer questionnaire is half art and half science. The iterative research and writing process can take months and usually involves three primary steps:
– hosting focus group to clarify issues and analytical requirements
-introduction and question writing
-testing and refining
When writing questions, simple and familiar language wins over slang, abbreviations and industry-specific jargon. If complex terms are unavoidable, define them for your audience.
Remember, copy for online or mail questionnaires won’t always translate well into other instruments like phone or in-person surveys. Write for all the types of surveys your audience will be taking.
4. Categorize questions properly
Within each survey category, group similarly structured questions together. For example, those that require a yes or no response, multiple choice questions, questions that require 1-10 rankings and open-ended questions should all be separated into sections. To pull respondents in, pose the most interesting or easy-to-answer questions first and save the more personal questions for last.
Finally, use the testing phase of the process to observe how focus groups respond to each question. Review any questions that respondents stumble over, consistently don’t understand or take too long to answer. Restructure these questions or scrap them altogether.
5. Consider using an incentive
Survey response is hard to predict and driven by many variables—most of them out of your control. Response rates can be improved by motivating your audience to respond. Depending on the nature of your business and your survey budget, consider entering each respondent’s name in a grand prize giveaway or offering small gifts or discounts. Though this makes the backend logistics more complex, it can often boost response rates significantly.
6. Manage the project closely
The importance of project management can’t be overemphasized. After development and testing, a survey needs to be administered properly. Choose a team member who’s highly detailed and focused on minimizing survey errors, optimizing responses, compiling data and presenting the results.
Analyzing the results
The same methods that worked in crafting the survey should be applied to presenting the findings. Pull the key pieces of information, present them clearly and show how they relate to the original survey objectives. Use charts and graphs whenever possible to present data visually. Then, encourage creative discussion around how the findings can be used to improve customer service, marketing, communications or employee satisfaction. Often, survey results can inform other parts of the business and the data can cross company lines to benefit other departments.
When it’s time to analyze the results, look for biases that are often unavoidable or hard to predict beforehand. Did only a certain set of zip codes respond to your mail survey? Was a particular e-mail domain underrepresented in your online survey? Why?
Surveys can be a valuable way to better understand any segment of a business, but they’re only as good as the strategy that’s put into them. When it’s time to launch your first survey, planning, preparation and the right people make all the difference.
(via OPEN Forum)