We are no strangers to random ads popping up on our phone screens or laptops asking us about our skin care or hair care regiment or our preference when it comes to certain brands or the kind of car we use and many such related questions every now and then. These surveys are a clever way for brands to gather consumer data.
Surveys are a great way to learn about existing as well as potential customers. They are not only cost effective but also really easy to implement. Surveys can be computer mediated or tailored for mobiles. Mobile surveys generally tend be more in-situ, in the moment rather than retrospective, meaning, they are usually presented when a purchase is being made.
To make an effective survey it is very important to carefully design the questions depending on the end goal of the survey.
Types of questions generally asked in surveys
1. Itemized Category
This category of survey involves a question along with a satisfaction scale which, in this case, has five options ranging from very satisfied to very dissatisfied.
Con: Consumer does not have a reference point to answer this question i.e., in this case, they do not have another fixed plan to compare it to. Most people might use their old health insurance plan as reference hence answers across people will vary.
2. Comparative Questions
This category of questions solves our previous problem. The metric scale in this case is comparative ranging from very inferior to very superior
Con: There is big loss of information in this method. For instance, both alternatives might not be great. Comparison of two sub-par alternatives does not serve our purpose as it gives us no real information about what customers really want.
3. Ranking Questions
This category of questions gives customers the liberty to rank their preference. The only rule being, no two characteristics can have the same preference or rank
Con: In the case of too many categories, this method renders ineffective. People might lose interest or get fatigued after carefully filling up 2-3 ranks. Number of categories should ideally be up to 6 and 8 at the most. If the number exceeds 8, all data after 2nd-3rd rank will most likely be compromised.
4. Paired Comparison
In this category of questions, what we are essentially doing is forcing people to pick one of the two alternatives. Based on their choice, we can extrapolate and understand what they like. This method is pretty effective since it closely mimics how customers act in the real world.
Cons: Customer might dislike both alternatives. Large number of brands cannot be compared by this method, it is typically used to compare 2-3 brands at a time.
5. The Likert Scale
This is the most common type of survey used by companies since it allows people to think about a lot of different statements at a time. It is extremely comprehensive.
6. Continuous Scale
This type includes a question, typically with a bar that slides either way on the scale to indicate the customer’s preference. It is generally used for more in-situ preferences. For example, this is commonly seen in election polls during debates when audience slides their bar to favor or support one candidate over the other.
All of the above-mentioned types have pros and cons. Hence, the type of questions to be asked in a survey should be decided based on the end goal, whether it is validity, reliability, customer satisfaction or other kinds of behavior.
留言