How to extract key consumer insights from open text answers

Did you ever want to hear your consumers’ real voice by including more open questions in a market research study?

A powerful solution is using synapse maps. Because a market research studies often sue “open” items, which allow for a greater variety of responses and new insights (e.g. “what will this product do to your skin?”) compared to closed, pre-defined questions. While clients love to hear from thousands of people, it is often hard to extract meaningful insights from such large datasets.

An example of typical survey procedure

In this article, we’re going to focus on open questions, and how to extract consumer insights by understanding how memory and associations shape the way consumers evaluate marketing material.

The main benefits of open questions are:

  • high information density
  • often elicit honest, unexpected responses
  • answers can include feelings, attitudes and understanding

The Challenge:

Although there are many benefits to using open questions, several pitfalls exist. Traditional methods focus on coding parts of participants’ answers according to categories relevant to the study. Although this offers a valid form of text analysis, it makes a study more costly and the insights are partly determined by the interpretation of the person coding the responses, introducing an additional source of bias.


Neuro Flash
2‘s Synapse Map© technology addresses these shortcomings by quantifying consumer’s’ opinion, wants and needs, using natural language processing (NLP). This makes results objective, quick to obtain, and free of many biases present in other methods. OurSynapse Maps are therefore a modern and unique way to visualize and unearth the hooks that appeal to most consumers, offering a glimpse into the way the brain processes and memorizes information.

 

The Test:

Suppose we wanted to test how two company names perform against each other in terms of first impression. For the purpose of this example, let’s call the first name “Neuro Flash” and the second name “Brain Flash.” The sample would be exposed to a short concept description, followed by the company name. At that point, we would split the sample into two groups and ask:

What are the first things that come to mind when you read the name [NAME]?

Half of the participants would be asked about Brain Flash, while the other half would be asked about Neuro Flash. Based on their responses to the open question, we would make a synapse map for each name:

aaeaaqaaaaaaaaimaaaajgeymjyxotaylwezyzgtndbhzs1iywu2ltm2zwi4ytzhmzvhma

Results:

Brain Flash:

Synapse Map Example Brain Flash

Neuro Flash:

Synapse Map Example Neuro FLash

So what exactly do these Synapse Maps mean? By separating components of the synapse map, we can look at various global trends.

Word Frequency

First, let’s take a look at how these names compare in word frequency:

             

Synapse map comparison

Brain Flash (left) Neuro Flash (right)

Bubble size indicates the frequency of words used in response to the question. The bigger the bubble, the higher the frequency. Line thickness indicates the strength of the association between words. The thicker the line, the stronger the association.

In this example, Brain Flash made participants think of “brain”, “fast”, and “big data,” while “neuroscience,” “ideas,” and “behavior” were the most commonly used words to describe the name Neuro Flash. Neuro Flash also elicited a more balanced response than Brain Flash, as shown by less variation in bubble size.

Significant differences

Colors on the Synapse Map indicate significance (p <.05), according to the legend below:

aaeaaqaaaaaaaaj2aaaajdyzzmzhndvllti5mzmtnguzyy05ymfkltvhnzvhywexyjfkoa

Brain Flash:

aaeaaqaaaaaaaafgaaaajgmxndq3ytczltvhowutnge1ny05yzvhltq0zjzjowu4ytizmw

Brain Flash did not elicit any significantly strong associations aside from the one between “good” and “idea . Noteworthy, however, is that the word “brain” was used most frequently when describing Brain Flash, probably due to exposure in the name itself. Additionally, Brain Flash elicited many significantly weaker associations between “science” and many other words, which goes directly against our intention.

Neuro Flash:

aaeaaqaaaaaaaaijaaaajdllode4n2q2lwizzdctndhjmy05odc2lwm5njjmyme3mdzjma
Here, you can see that the name Neuro Flash elicits a strong, significant association between “neuroscience” and “behavior”, while it elicits a significantly weaker association between “perception” and “behavior.” This is representative of our company’s goal, using implicit methods based on neuroscience to predict behavior and relying less on consumers’ perception to predict behavior. Thus, the name Neuro Flash fits the company’s identity based on the first impression it elicits. Neuro Flash is thus, the clear winner.

Conclusion:

As it is clearly seen in the example, the visualization of the data is self-explanatory and provides a straightforward way of conveying the main themes in the responses obtained from open-ended questions. By analyzing a variety of open-ended items in this manner, we are able to reconstruct the thought processes behind each element tested. Critically, adding more open questions can be done with minimal impact on study costs and timing. 

aaeaaqaaaaaaaaksaaaajdlizwy0nmq3lwiyzgitngiyzc1iyzq5lwi1mgjkngm1mdhmyw
Seeing words and associations that match the concept and brand vision, leads to better decisions. 

Thus, Synapse Maps are a useful technique presenting data from open questions. The visual style allows a clear view on the main themes and makes it very easy to understand how ideas are formed and developed in the consumers’ minds. Furthermore, our Synapse Map technology can be applied to all open questions, providing results across a range of key performance indicators (KPIs).

Questions or comments? Get in touch with me.

References:

  • Erickson, P., & Kaplan, C. (2000). Maximizing qualitative responses about smoking in structured interviews. Qualitative Health Research, 10(6), 829-840.
  • Friborg, O., & Rosenvinge, J. H. (2013). A comparison of open-ended and closed questions in the prediction of mental health. Quality & Quantity: International Journal Of Methodology, 47(3), 1397-1411. doi:10.1007/s11135-011-9597-8
  • Huang, Y., Chen, C. H., Wang, I. H. C., & Khoo, L. P. (2014). A product configuration analysis method for emotional design using a personal construct theory. International Journal of Industrial Ergonomics, 44(1), 120-130.
  • Petiot, J. F., & Yannou, B. (2004). Measuring consumer perceptions for a better comprehension, specification and assessment of product semantics. International Journal of Industrial Ergonomics, 33(6), 507-525.
  • Rasoulifar, G., Eckert, C., & Prudhomme, G. (2015). Communicating Consumer Needs in the Design Process of Branded Products. Journal of Mechanical Design, 137(7), 071404.
1 reply

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *