Getting Started with Surveys

October 18, 2022

Getting Started With Surveys

Surveys can be an extremely useful tool for gathering information from individuals and groups. They are used across disciplines and industries to help researchers learn more about populations and gain actionable insights. I’ve done extensive survey research in nonprofit and industry settings, using these data to improve programs, make changes to technology, design communications plans, make content acquisition decisions, and much more. In this blog post, I’ll focus on one of the most important parts of survey research — planning.

Broadly, surveys fall into three categories:

  • Quantitative, with multiple choice questions or numerically ranked items

  • Qualitative, with open-ended questions, and

  • Mixed-methods,a mix of quantitative and qualitative survey questions.*

Each method has its own strengths and weaknesses, and no technique is 100% fail proof. 

One of the most important parts of survey research is planning. If you aren’t thoughtful about why you want to use a survey to learn something (as opposed to other research tools and methods), you might spend a lot of time, energy, and resources — financial and otherwise — building, fielding, and analyzing a tool that won’t actually help you accomplish your goals.

Before getting started, here are some key things to consider:

  1. Why do you want to do a survey? Surveys are a great way to get people’s opinions, quantify trends in a population, and get answers to open-ended questions. But is a survey the right way to get the information you’re seeking? If you want stories and experiences, consider interviews or focus groups. If you want to test your technology or new features of a product or service, you might want to explore user experience testing. Just because other companies or research teams are fielding surveys doesn’t mean they are the right tool for your project.

  1. Who are you trying to survey? Are you trying to reach a nationally representative sample of people? A specific group of program participants? The potential audience for a new television show? A very specific niche group, like unicycle enthusiasts in rural areas? Thinking hard about who you want to learn about or from will help you determine if a survey is the right way to reach this demographic and get you thinking about your sampling strategy.

  1. What exactly are you trying to find out? It is extremely important to figure out what this survey is designed to uncover. Are you trying to measure satisfaction with an experience, the impact of a program, if there is a potential market for a new product, or something else? Write out your objectives so that they are very clear (and for your team to agree upon).

  1. How will you use your survey results? After the survey is fielded and analyzed, how will you use what you learn? Being able to articulate the linear relationship between your objectives and how you will use new insights will help shape the survey will ensure that the survey tool will get you the answers you need.

  1. How many responses do you need? Do you need a large enough sample to test for statistical significance at 95%? Are you looking for 10% of an existing user base? This will be important for calculating your budget, establishing a timeline, planning the scope of work, and making sure your data is ready for whatever testing you and your stakeholder need.

  1. When do you need answers? Surveys take time! Developing a survey tool and getting it into the field can take weeks, not to mention the time needed for analysis afterward. Understanding the timeline of a survey can help you shape the survey tool and your expectations.

Creating an effective survey tool is both an art and a science. By spending more time in the planning phase and thinking about these questions, you will be more prepared to build a useful survey tool, enter the field intentionally, and analyze your results effectively.


* Ponto J. Understanding and Evaluating Survey Research. J Adv Pract Oncol. 2015 Mar-Apr;6(2):168-71. Epub 2015 Mar 1. PMID: 26649250; PMCID: PMC4601897.