This is a very common question people ask before sending a survey. You can follow these steps and the chart below to determine how many people you need to interview in order to get results that reflect the target population as precisely as needed.

Identify the size of your target population

If your entire target audience is 2,000 people or fewer, send to as many of them as possible. If the target population is at least 2,000 people, consider sending to a subset. Use the information below to identify how many responses you need.

Determine the accuracy you’ll need

You’ve heard a “plus-or-minus” figure reported along with polls. This margin of error, or confidence interval, indicates the accuracy of the data. If a survey result is 60% with a 5% margin of error, that means the true value lies somewhere between 55% and 65%. Confidence level is the next measurement that will impact the number of respondents you need. This measures the certainty that the survey results are within the margin of error. Most researchers use a 95% confidence level. When you put the confidence level and the confidence interval together, you can say that you are 95% sure that the survey results are within the margin of error (between 55% and 65% in the example above). The wider the confidence interval you are willing to accept, the more certain you can be that the whole population answers would be within that range.

Estimate the number of completed surveys you need

The chart below allows you to estimate the number of completes you need when you use a 95% confidence level. Select your approximate Target Population and choose your margin of error.

Note that once your target population exceeds 10,000, the number of additional respondents needed grows quite slowly.

If there is a segment of your target audience that you would like to analyze – such as companies of a certain size for customer satisfaction – aim for a minimum of 75 respondents among that group. For example, if you know that large companies are 40% of your total customer base, then you will need approximately 200 total survey respondents to get 75 large companies. If you weren’t planning on surveying at least 200 people based on the chart above, you would need to increase your number of completes in order to analyze the large companies.

Send the survey

Remember, you’ve determined the number of completes you need. To estimate the number of survey invitations you need to send, divide the number of responses you need by your estimated response rate. Survey response rates vary based on your relationship to the survey recipients, survey length, topic, etc.

You can find more information on response rate, and tips for improving yours, in the posts “What is a typical response rate for a survey?” and “12 Tips To Help Increase Your Survey Response Rates”

Thanks for the information – much appreciated

Michael – Thanks for your comment, I’m glad the information was helpful to you.

Have you ever considered publishing regular survey tips to your blog – maybe one every day to keep people fresh.

Dave – Thanks for stopping by and leaving your comment, I have passed your comment along. We do publish a Survey Tip of the Day on Twitter so you can follow us there if you would like.