How to design a good and effective survey for your business?

by Ashvini on April 9, 2011 · 5 comments

in Business

A survey is an important tool to get feedback from customers, employees, new markets segments etc.

Surveys are often used by market researchers to understand the trends of the local market. A survey is just a set of questionnaires that are designed to elicit a particular response that can be categorized and cataloged for further study and research.

Following are the types of surveys that can be the most popular for business and companies.

  1. Employee satisfaction surveys
  2. Customer Satisfaction surveys
  3. Market research survey

A survey may be conducted using multiple channels. They are

  1. Asking in-person questions
  2. Asking people to fill up forms
  3. Using web forms on internet
  4. Asking questions on telephone

It takes usually a lot of analysis and thoughts before a survey design gets to see the light of the day. It is important to know that a badly designed survey will not only cause the survey takers to lose interests but also it will make gathering right data difficult. Further difficulties include incorrect analysis and wrong trend predictions.

So it is very important that mistakes are avoided at the start of design of survey at least.

Of course, it is not possible to have a perfect survey design at the first itself, but every survey campaign will have a lesson to be learned.

Here are some of the things you can use and some of them to avoid while designing your surveys.

  1. A survey that asks people to reveal their personal information too much is a no-no
  2. While it is true that a survey should try to get as much information as possible, those that ask a responder to reveal too much personal information are not going to gather much information. People are normally wary about giving out personal  information. It has been seen that mobile number that were asked in the survey were later used to send unsolicited mesages( for e.g. marketing messages ).

    If you are asking people to reveal their personal information, provide a line or two about, the manner in which you are going to use this information. If people have trust in you or your brand they might choose to provide that information. However making personal information mandatory in a survey is a sure short way for disaster.

  3. A long 100 question survey is again a no no
  4. If you have a survey that is goes into pages or looks like a government form, be assured that no one is going to fill it. A short survey keeps the attention span in limit and provides best answers. A survey normally should not take more than 5-6 minutes to fill and easy to understand is most favored.

  5. Posing the survey to wrong segment is going to get you wrong answers
  6. Question made for kids should be answered by the kids only. If you are going to ask questions to a mall going crowd to people standing in the flea market, you will probably not get the right answers. It is very important to know in advance what is the segment you are going to target and how you are going to find it,

  7. Having a survey with confusing answers
  8. Many a times answers to survey questions are ambiguous. While there are areas where clear cut line between two opinions cannot be achieved, it is important to make survey answers as unambiguous as possible. If the question cannot have ambiguous answers then it is a good idea to separate them into different questions.

    For e.g. a non ambiguous question is
    Whom are you going to vote for this year? Republican or Democrat
    Slightly ambiguous
    If your chairman gives you free classes to attend would you vote for her? ( Many people might not want to get into free classes but would look for other parameters before voting for chairman)
    Totally ambiguous
    How much happy you say you are ? (Open ended and ambiguous). Here the responses may vary from not at all to very happy. This type of question may be easy to answer but is very difficult to compile and collate.

    So the idea is to have a question that has mutually exclusive answers or provides enough options to people to chose from.

  9. Asking wrong questions to the right audience
  10. Even if you have found the right segment, it is quite possible that the questions that you are going to ask is now irrelevant for the responders.
    For e.g. if you ask a couple who are unmarried or just dating about family planning, you are asking them a wrong set of question.
    Or if you asking people’s opinion about some vague rule in a country that they have not heard of, you are again asking wrong questions.
    Or  you are asking people about a new idea which they have no inkling of.
    Asking people about things, they have not thought or heard about at all, will result in vague unusable response. Ask relevant questions to get the right response.

  11. Making the questions very long
  12. People are normally in a hurry even if you manage to catch them . If your survey questions are very long then people will either not listen to them that well or they would be inclined to finish the survey fast. In that case, they would give answer without pondering on them too much . It is thus important to keep survey questions short and crisp so that they capture users attention.

  13. Asking negative or biased questions
  14. One of the real interview in media went like this. In a family a man was murdered. Media went to the house to enquire about family. The first question to grieving relatives was “How are you feeling?”. It is a real incident and I am not making it up. It was one of the most stupidest question to be ever asked. It was insensitive and callous.
    I am pretty sure not many will make such a mistake. However it is a good idea to review your questions and weed out any insensitivity that can turn off a responder.
    Asking questions like “Do you think factory staff should be paid more ” is also confusing and negative in nature and reworded to get right response.

  15. Using jargon in the survey questions
  16. Avoid jargon as far as possible. This may include words that are used by youth ( on sms’es for e.g.),management words such as black belt, six sigma etc or technical terms. If you do need to use them, then provide a basic explanation of terms that you are going to use later.

  17. Using difficult font size, unattractive layout, confusing navigation
  18. It is very important to have the right layout which is easy to navigate and easier to read. A layout where lot of questions are crammed or which jumps from pages to pages without a logical consistency is going to turn off a responder. Use basic colors for background and a high contrast between background and foreground.

  19. Have consistent responses
  20. If you use yes/no type of answers, make whole survey a yes/no type of survey. Otherwise use any other consistent system that does not change the responses with every question. If yous responses are consistent then it will far easier for readers to read the question and just put a tick wherever they feel is right.
    Changing response type frequently is surely going to turn them off.

About The Author


aderonke adebunmi January 17, 2013 at 5:14 pm

i find it a little bit difficult to construct a questionaire based on family disorganization……….what can i do.

Kyle October 13, 2011 at 8:49 pm

Thanks for the advice on how to create an effective survey.

Business Logo Design April 11, 2011 at 12:05 pm

Fabulous man! What an incredible article you have posted. I have to confess your great effort. Such a captivating stuff!

Logo designer June 22, 2011 at 6:03 pm

A good and effective survey is very important for the business. Such surveys help the businesses to know more about customers , market and the competition. Thanks for highlighting this important aspect of business.

Aswain April 11, 2011 at 10:24 am

Hello Ashvini, great post again. I have never created any survey till date but I have participated in few of the surveys, very recently in a survey conducted by Microsoft. It was a good experience.!

btw, I have often found most of the surveys quite complex and at the same time very boring. To me, a survey should be short and should be interesting enough to make people participate in the same. You have explained it so well in your points. Keep it up!

Comments on this entry are closed.

Previous post:

Next post: