One of the most important challenges that we face at Openmet when it comes to customer surveys is the need for brevity. In general, the time spent by customers completing surveys tends to be very limited, and we have to make the most of that attention span to obtain as much feedback as possible.
This time limit results in serious constraints on the survey’s complexity and number of questions. Among other techniques used to minimize these constraints are the need to remove superfluous questions or those whose answer can already be inferred or known in advance (for example, thanks to the customer data held in the CRM system), or the need to focus on the most important and relevant aspects of the survey.
In spite of all our efforts, with more questions than we can fit in the space and time available. In the case of surveys with a sufficiently large sample, question rotation is one of the most effective techniques to help us minimize this problem.
How do surveys with question rotation work?
If you think of a survey as a jigsaw puzzle, you can split your ideal questionnaire (the whole puzzle) into small units (the pieces). Surveys with question rotation do exactly this:
1. Based on a pre-established algorithm, you choose a small portion of the questions comprised in the full questionnaire (you choose a piece of the jigsaw). This piece will change (rotate) for each customer.
2. The customer is asked only the small selection of questions chosen.
The logic used (the algorithm) to choose which questions to include in each piece of the puzzle can change according to your needs:
• You can include a question that is always fixed and others that change.
• You can decide to add questions based on individual questions or on pre-established blocks of questions (which cannot be split up).
• You can decide which questions (or block of questions) to include either randomly or sequentially.
An example of this could be where you have a survey of 20 questions of which we ask customers 5 questions in rotation. In that case, if you can obtain answers from 3,000 customers, you will end up with 750 answers for each of the questions in the survey.
By way of conclusion, surveys with question rotation are a very good strategy for occasions in which:
• You have many questions but you know that the customer will have a short attention span. Examples include pop-up surveys on web servers, automatic telephone surveys, surveys at POS kiosks, etc.
• The structure of the questionnaire makes it possible to split it up.
• The samples are sufficiently broad to ensure the volume required to obtain enough answers for each question.