If you’re in marketing or customer experience, you know how vital customer sentiment towards your brand is. In an ideal world, you would know what each client is thinking about your brand and customer experience, and how they perceive you against competitors. Since you can’t read minds the next best thing is to collect data via surveys.
Engaging surveys can bring in honest and candid responses from surveyees and can provide valuable insight into your company’s customer experience department and brand reputation. However, you likely don’t have the time or resources to survey every customer—particularly if you’re a start-up or small-to-medium business.
Therefore, you want an ideal sample size to closely encapsulate and represent your entire customer base. Plus, you want to avoid response bias. To understand your ideal sample size for surveys you need to know these terms first. Let’s go back to your high school or college statistics course.
- (N) Population size is the total number of people in a specific demographic that you want to measure.
- (e) Confidence interval, or margin of error, is the percentage of error you should be willing to allow in your results. Usually portrayed as a plus or minus percentage range (+/- 3%).
- Confidence level means the percentage of how sure you are that your results fall within your confidence interval. Typically, the confidence levels are 90%, 95%, or 99%.
- (P) Standard deviation is the estimated percentage of deviation you can anticipate among your survey results. Note, standard deviation is difficult to calculate before running a survey. Therefore, use 50% or 0.5 in your calculations to play it safe.
- (z) Z-Score symbolizes the number of standard deviations between a selected value and the average number of the population. In other words, it means how your results compare to a “normal” population.
Here is a quick graphic and sample size equation explaining how these different terms work together to calculate the sample size.
If you’re wanting a quick way to calculate sample size, then use this sample size calculator offered by Chattermill. Lastly, note that in general, larger sample size will lead to more statistically accurate results.
For more tips on how to create the best online surveys that will help grow your business see Chattermill’s infographic below: