Test your survey
Table of contents
Before you actually distribute your survey, we recommend that you do a test first. After putting in the time and effort to create a survey, you want to make sure that it works and looks the way you intended—for your sake and for your respondents’.
Testing allows you to pick up on issues like broken logic, formatting flaws or navigation issues. When testing, the worst-case scenario is an error message, but the implications could be much worse if issues are not picked up before the survey is made live.
With this in mind, you should test each survey from start to finish before final release. Should there be an error or other issue in your survey that impacts the outcome of your results, the only solution would be redistributing the survey or accepting the less-than-ideal results.
As part of the testing process, we recommend sending a testlink to a few colleagues so that they can test the survey and give their approval before publication.
How do I test my survey?
To test your survey, there are a couple of possibilities depending on where you are in the creation process and what you are interested in previewing.
- Preview a page
- Preview the entire survey
- Send testlink
- Send a test sendout (email or SMS)
This option is available on the Distribution page of the creation process. If you scroll past the different distribution options, you will see a box called “Distributions”. In addition to seeing your selected distribution methods you will see a button called Testlink on the right side.
By clicking this button, you create a URL and a QR code that can be used to test the survey.
Using a test link will give you an authentic version of the survey with the possibility of testing logic and jumps, as well as translations (if you have any). Test links do not register as survey answers—they are simply a means to test the functionality and appearance of your survey. For this reason, you should never distribute a testlink to genuine respondents, as their answers will not be saved or accessible.
When you are working with an E-mail or SMS sendout you can test the entire respondent experience by sending yourself or others a test sendouts. This can be useful to make sure that you are satisfied with the format of the e-mail or SMS that is sent, and that there is a working link included in the message.
When you are inside an e-mail or SMS sendout, click the button with the paper plane on the right side(e-mail) or bottom (SMS) of the form.
What to include in the sendout text?
- What is the survey about?
Be clear when outlining the content and purpose of the survey. Try to keep this as concise as possible.
- Why should the recipients respond to the survey?
Your response rate can increase if you incentivise your respondents to complete the survey. You can do this by defining a clear purpose for participants and explaining how vital their responses are to improving your offering.
- How long will it take to respond?
Be honest! If you are unsure, ask some colleagues to respond to the survey and see how long it takes. If it is a short survey, you can use that as an incentive to respond and if it is a longer one, it is good to provide the recipients with plenty of motivation to respond to the whole survey.
- How will the results of the survey help you?
Do not forget to let respondents know how important their feedback is and how it will help you to improve your product/support/service.
- Is the survey anonymous?
If respondents will remain anonymous, be sure to tell them as this could encourage more honest answers.
- How should the subject field be worded?
The subject field is the first thing that a respondent reads. Keep it short, easy to understand and with a clear purpose. You can also do a test by sending two e-mails with different subject fields to a small group of respondents to see if either of the emails receives a larger percentage of answers.
- Which email address is the survey sent from?
Use an email address that the respondent will immediately recognise, otherwise you will risk having the e-mail dismissed as spam.
- Who is the sender of the survey?
Most companies use the company name as the sender, but it can also be effective to use a specific person as the sender. You could, for example, send the survey from the product manager if the survey is about product development or from a sales representative who has had direct contact with the customer.
When is a good time to send a survey?
The answer to this question is not clear cut. It depends on the target group and any circumstances that may affect their ability or desire to reply, e.g. public holidays or time of day. Generally speaking, sending surveys early in the week seems to get the best response level.
This graph is based on Netigate data about the number of responses received on different days of the week: