Expert en feedback management - Spécialiste dans la mise en place de baromètre de satisfaction et d'enquête de satisfaction
Expert en feedback management - Spécialiste dans la mise en place de baromètre et d’enquête de satisfaction
Votre demande à bien été prise en compte.


10 September 2011

Some rules when creating your customer satisfaction survey…

1) Ask questions in a structured way

When creating your customer satisfaction survey, it is just as important to have a structure for yourself, as for your respondents.

For yourself because it ensures you never miss a key question.

For your respondents as there is nothing more frustrating than not understanding the logic behind a customer satisfaction survey.

For customer satisfaction surveys, the order of the questions should generally follow the logic of the customer experience.

Here’s an example for a store purchase:
- Location of the store
- Layout of the departments
- Helpfulness of the sales staff
- Paying at the till
- After-sales service

As in the example above, you must also group questions by attribute of satisfaction.

2) Keep questions to a minimum

Which questions will actually be of use to me ?

Many of our clients use their customer satisfaction surveys to ask all kinds of questions, out of curiosity (or blood-mindedness?) without really worrying about how these will be analysed later on. We therefore see cases where dozens of questions are asked when in the end only 2 or 3 indicators are followed up.

Are the questions asked really related to customer satisfaction ?

A customer satisfaction survey has only one goal: to measure customer satisfaction. Sounds simple but it is not because a question ends with “On a scale of 1 to 10” that makes it a question of customer satisfaction is understood, relative to the customer experience whose satisfaction you want to measure.

If that seems difficult to surmise before the fact, (except for obvious cases), it is easy to measure after the fact. To find out, it is usually enough to calculate the correlation coefficient between the questions asked and the overall satisfaction relative to the customer experience.

Are the questions you ask relevant ?

Are the questions you ask relevant ?

Once again, it seems difficult to surmise beforehand… but when we analyse the results, we look at a very simple indicator: variance. If, on average, respondents vary little in their ratings to a series of questions (always on the same attribute) then it is very probable that the number of questions is too high or that the questions are not relevant enough.

Retour à la liste


Fabien HOURI
Operations, Associate

Do you need more information ? Contact the author :

Understanding customer satisfaction

Logiciel sondage – Son intérêt pour votre sondage de (...)