The Future of Marketing Research: Text Analytics

While a long-term topic of discussion in the marketing research world, there does seem to be general agreement that it’s not a question of which is better, but rather that both are valuable for different kinds of information. The challenge has long been in interpreting the qualitative, or verbatim information, especially with the structure of a quantitative survey.

There are a lot of challenges to including open-end questions in survey research. Most importantly, respondents don’t like to answer them and tend to give incomplete answers to open-end questions in online or paper surveys. Telephone surveys, of course, have the advantage of being able to have the interviewer probe the respondent to gain a complete answer, but researchers are completing fewer and fewer telephone surveys in favor of online methodologies. Secondly, responses are difficult to interpret. Hand-coding each open-end response can add significant time and expense to your survey process.

However, there are seven great reasons to use open-end questions in your surveys:

  1. Get All Possible Answers. When designing close-ended questions, you might not know all the possible responses that fit each respondent’s situation. Researchers usually solve that dilemma by including an “other, please specify” response, but that may still not give you all the insight you need. Open-end responses allow the respondent to tell the researcher anything they believe to be relevant and responsive to the question, without any prejudgment on the part of the researcher. And the bonus is that they tell you their answer in their language, as well.
  2. Get the Detail. In addition to any possible answer and using the respondent’s own words, open-end questions allow respondents to include as much detail as they like. That extra detail can help you better understand not only their answers but how they perceived and understood your question.
  3. Uncover the Unexpected. If you knew all the possible answers, you would make it easy on the respondent and the analyst and use a close-end. With open-end questions, you get responses you may not expect. And these unanticipated answers can be surprisingly valuable, whether in identifying new and innovative products or in crafting more effective marketing messages.
  4. Complex Answers to Complex Questions. Sometimes, an open-end question is the only one that will work. If the answer is dependent on context, or on an individual’s information, there may not be a way to scale a close-end question appropriately to yield sufficient information for analysis.
  5. Creative Self-Expression. Open-end questions let respondents surprise you in their responses. Whether it is a particularly apt turn of phrase, a perfect simile or image, or just a very eloquent response in the customer’s voice, open-end questions give you the opportunity to uncover additional value.
  6. Understand the Thought Process. Sometimes the specific answer is less important than the journey the respondent takes to arrive at their answer. By understanding how the respondent thinks about the question and the pieces of information used in coming to an answer, you may learn valuable lessons about communicating with and serving customers.
  7. What You Don’t Know You Don’t Know. It’s widely recognized that we know what we know, and we know what we don’t know. What trips us up is what we don’t recognize we don’t know. Open-end questions can help you avoid that and can open the door to a new perspective on your customers.

Open-end questions and the verbatim responses they generate are very important and should be included in every survey. However, we shouldn’t stop with open-end questions in surveys. Anytime we can access what a customer says or writes about our brand, product, or industry; we should consider that open-end information. So open-end information can be retrieved from reviews, texts, social media posts, in-bound chats or phone calls, blogs, vlogs – just about any communication from the customer. (And don’t forget the speech-to-text tools that will give you access to voice recording information as well.) In fact, communications that are not driven by or to the business might be even more revealing due to the lack of potential bias.

But if we are going to use open-end responses – and even expand it further to any written communication, we are going to need a faster, more efficient, and cost-effective way to analyze and summarize this information. Which is why text analytics will be an important tool for marketing researchers soon.

Text analytics is, according to the CX Dictionary, “the process of drawing meaning out of written communication. In a customer experience context, text analytics means examining text that was written by, or about, customers. You find patterns and topics of interest, and then take practical action based on what you learn.” While this process can be completed manually, it is very inefficient. Recent advances in text mining and natural language process algorithms make the process faster and more productive.

Text analytics has evolved tremendously over the past few years, and its evolution will continue to make these invaluable tools resources for marketing researchers. For now, start by adding open-end questions into every survey as an opportunity to learn the unexpected. And keep an eye on the evolution of text analytics as it develops for ways your business can benefit.

Leave a Reply

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