Survey completion rate vs Response rate

Terms like response rate and completion rate are frequently used by marketers and surveyors. Because of their similarities, it is quite easy to confuse the two. Here we will try to understand what the two terms actually mean and how they are related—but actually quite different—when it comes to surveys.

The key similarity between the response rate and completion rate is this: the higher the rate, the better the survey's results and outcome.

So let's dive into the two concepts and see how they influence your survey's sample rate and statistical accuracy.

What is completion rate?

An online survey's completion rate is the number of people who have fully completed the survey divided by the number of people the survey has been sent to. If a person has started but not completed the survey, they will not increase your completion rate.

How is completion rate calculated?
What causes a low completion rate?
- Your questions are not properly framed
- Your survey might be too long
- Tedious and/or ask personal and sensitive information
- Misleading to respondents
- Not apply to your chosen population
What is the effect of a low completion rate?
How do I improve my completion rate?
- Keep it simple
- Keep it short
- Keep it easy
- Keep it crisp

How is completion rate calculated?

Let's look at how the completion rate is calculated using the following data:

Number of emails sent: 500

Number of people who opened the survey: 250

People who completed the survey: 200

Completion rate = No. of surveys completed / No. of respondents who opened your survey

So, in this case...

Completion rate = 200/250 = 0.8

Completion rate = 80%

The thing to notice about the completion rate is that it only depends on the people who have interacted with your survey, i.e, who have at least opened it. How many people the survey was sent to is not taken into account in this case.

What causes a low completion rate?

A low completion rate is an indirect message to you from your respondents that they do not like your survey. Some of the reasons for this could include the following.

Your questions are not properly framed

In this case, you need to ensure each of your questions has a proper agenda.

For example, if you want to ask a question about employee health because you're trying to find a correlation between shift timings and snack timings, it would be a better idea to ask those as separate questions.

What shift are you in and what time do you usually snack?

This question is poorly framed because it throws two different questions to the respondent at once: their shift time and their snack time. These may not necessarily coincide for the employee.

So instead of asking two questions at once, it's better to ask them as separate questions simply:

What shift are you in?

When do you usually snack?

Asking the two questions separately does not confuse the respondents and makes it easier for you to assess the correlated result in the survey.

Your survey might be too long

Try not to make your survey unnecessarily long, with too many questions, as this may bore respondents and force people out of your survey.

For example, a good number of questions to go with for a mid-sized survey would be around 10.

Your questions might be tedious and/or ask personal and sensitive information

While writing surveys, it is better to avoid topics like religion, community, creed, gender, and other sensitive issues that people may not be as willing to answer or that might be likely to result in a biased response. However, if your survey does deal with sensitive issues, assure your respondents that their data will be protected.

Your questions might simply be misleading to respondents

Avoid phrasing questions in a way that egg the respondent towards one part of the question. These kind of questions are called leading questions and often confuse the respondent.

Example:

How short is the Eiffel Tower?

Instead of assuming that your respondents think the Eiffel tower is short, it's better to frame it like this:

What do you think about the height of the Eiffel Tower?

In this manner, you are not assuming anything about the respondents.

Your questions might not apply to your chosen population

Ensure you have chosen the right set of the population to ask questions from. Asking questions about the performance of a smartphone brand to population over the age of 60 may not always be the ideal way to go. In this case, you should go for a younger target group as they are likely to be more tech -savvy.

Always double-check your questionnaire to ensure you have not committed any common question errors. If you do not have any prior contact with your survey respondents and do not know if they are the right audience for your survey, you can always buy responses from the exact audience you want.

What is the effect of a low completion rate?

When people are dropping out of your survey before completing it, they are not filling out the information you need. This implies that answer data from some questions will be more reliable than others.

For example, if some of your respondents are dropping out before answering the last question of your survey, data from this last question is not as reliable, because there are fewer people answering it.

How do I improve my completion rate?

It is tough getting people to answer your survey completely. Here are some simple tips you can follow to ensure most of your respondents answer all questions.

Keep it simple

Always start your survey with a simple, easy-to-answer question. Surveys that open with an easy multiple choice question have a higher chance of keeping respondents going. By comparison, surveys that begin with an open-ended question have a significantly higher chance to be left incomplete.

If you do have an open-ended question, make sure not to ask it in the beginning—that way, you allow the survey taker to get into the rhythm of asking easier questions first.

Keep it short

Research says that longer surveys have lower completion rates because people simply get tired and withdraw. Remember, when you are making people answer survey questions, you're asking for their time, and you need to be mindful of it. As we mentioned previously, a 10-question survey is a good length if you are aiming for a higher completion rate. A good tip is to lay out a survey goals plan where you assign each question to one goal. Don't ask for unnecessary information you do not really need.

Keep it easy

We know, there are times when you simply have to ask a matrix question, ranking question, or even those with a text box. That is perfectly okay—asking one of these questions at the end of your survey will not do much harm: but do restrict it to only one.

Keep it crisp

Each additional word in your questions can negatively impact your completion rate. Just like long texts generally tend to tire people out, so do long survey questions. Even if you do have a question with a lot of text, make sure not to ask it at the beginning. However, do not sacrifice words for the sake of clarity.

For example, if you are going to ask a question on fitness trackers, there can be two ways of framing it:

Are you going to buy a fitness tracker?

This is not as clear as asking it like:

How likely are you to buy a fitness tracker like Fitbit or Mi band?

Please keep in mind: Do not over-correct the length and sacrifice question clarity.

To sum up the mantra: To get good completion rates, keep your survey questions simple, short, easy, and crisp!

What is Response rate?

Response rate is a statistic deceptively similar to the completion rate. However, it manages to provide a deeper insight into survey results than completion rates. Simply explained, response rate refers to the number of people who completed your survey divided by the number of people who comprise the total sample group.

How is response rate calculated?
What causes a low response rate?
- Wrong sample
- Lack of a good branding
- Non-response bias
What is the effect of a low response rate?
How do I improve my response rate?
- Run a pilot survey
- Respect your respondents
- Offer surveys through multiple channels

How is response rate calculated?

Let us use the same example as before to explain the response rate.

Number of emails sent: 500

Number of people who opened the survey: 250

People who completed the survey: 200

Response rate = No. of surveys completed / No. of people in your total sample group

So in this case, our calculations look like...

Response Rate = No of surveys completed / No of emails sent

Response Rate = 200/500 = 0.4

Response Rate = 40%

What causes a low response rate?

The reasons your survey has a low response rate can depend on many things. Here are some of the common ones.

Wrong sample

Low response rates may be an indicator of picking out a sample that is simply not interested in taking your survey. In this case, you have to research the target segment for your survey properly and then pick them out.

Assure the people you select are really the right ones to ask those questions. You have to ensure you have chosen the right sample for your survey before sending it.

Lack of a good branding

The outlook and the overall manner of how you present your survey is quite important. Always find a way to entice your participants. Ensure that the subject lines are crisp and branded properly to instill a sense of credibility. You could also use an extra incentive to drum up something useful from your desired sample.

Non-response bias

Non-response bias occurs when people in a certain demographic—say gender, age, or social class—are not taking your survey. There could be many reasons for this: you sent the invite during a religious holiday, an email browser is labeling your invites as spam, or your topic is a sensitive issue for some people.

What is the effect of a low response rate?

A lower response rate could potentially damage your survey in that your original sample becomes smaller and smaller. This is awful news for your margin of error and the reliability of your survey reports.

For example, let us consider we have 300 respondents for a sample population of 1000. As per our margin of error calculator, even if you manage to get a 100% response rate, you will still end up with an industry standard margin of error of 5%. Since a 100% response rate is not always a reality, we have to lower our response rate. Let us consider a 30% response rate (which is quite high itself by survey standards). In this case, our margin of error inflates to 10%, so this highly affects your survey accuracy.

How do I improve my response rate?

There are many factors that influence your response rate: your target population, relationship with survey respondents, the survey invitation, survey length, question framing, survey topic, and incentives, among other examples. While you may not have complete control of all these factors, let us look at some over which you do:

Run a pilot survey

The trick to eliminating non-response bias is to conduct a pilot survey before sending it. A pilot survey is a test run for your survey that can help you filter any errors that you may be making. Undetected bias creeping into your survey can skew results and hurt data credibility, so it is important to be aware of the kinds of bias like response bias and confirmation bias.

Respect your respondents

To get people to want to complete your survey, ensure you express your appreciation for their participation. Incorporating phrases like "We want to know what you think" or "Please take five minutes to complete your survey" may feel minimal, but they go a long way in propelling people to take some time and fill out your survey. Be respectful of their time—often giving an idea of how long it will take to answer the survey at the beginning also helps.

Offer surveys through multiple channels

You may have your respondents' email addresses, but that does not restrict you to sticking to only email surveys. Try posting your survey on social media channels or embedding it into relevant marketing pages or via an SMS. Often the second or third time it comes to their attention, they may complete it.

Now that you are aware of the differences between the response rate and completion rate, you can start viewing your survey success rates with a fresh lens.

Start surveying away!

Survey completion rate vs. response rate: Differences explained