How to find people to answer your market research survey

Author: Jim Whaley

What are surveys and why should you use them

If you are starting a business or have an established brand already chances are you have either thought about surveys or have many of them under your belt.

Surveys are a great way to gather a broad range of information about your products and services and specifically how people relate to them. Surveys can answer questions about a wide range of topics and issues that impact your business and your specific products:

  • Market share and market sizing
  • Market segmentation
  • Product concept testing
  • Attitude and usage study
  • Pricing study

    • While these are just a few examples of surveys where reaching customers and prospects is important there are many more types of surveys are being done around the globe each week, continuously 24 hours a day.

    Public opinion shapes the way many important social changes occur from public transportation and services to public health and welfare. This is why it is important the researcher design their survey to include the opinions from a representative sample of the population.

    How to design online surveys to ensure a representative sample

    Unless you are surveying from customer lists or lists of any other kind, chances are you are not going to have the opportunity to send an invitation to every person on that list. This means you will likely to be turning to a market research panel to reach your target population.

    Depending on the target population you are trying to study your sample provider can help determine the right options for targeting your audience and the right research sampling methods to use for your situation.

    Online surveys which are quantitative in nature should be drawn from a random sample of the target population in a manner that every member had an equal chance of being selected. This is otherwise referred to as a probability sample and the results of which a study can have statistical inferences drawn to make conclusions about parameters of the target population

    An example of such a conclusion would be if you conducted a U.S. Nationally Representative General Population Poll of N=1000 people and 45% said they would vote Trump and 55% Clinton. That is how the overall popular vote would come** in if it were at the U.S. level and the Electoral College was not a factor. (This article has no political bias either way but rather it is for demonstration purposes)

    • Margin of Error +-3% and Confidence Interval of 95%