ProjectBlog: What we learnt from Ray Poynter’s sense-making with open-ended text

Blog: What we learnt from Ray Poynter’s sense-making with open-ended text

No company can ever really win the market without actually understanding the difference between what the customer thinks she wants and what she actually wants that’s why open-ended text plays an important role.

Why do people buy sedans when they love convertibles?

Some years before, there was a study by Dr.Dichter on why car companies showcase convertibles on the windows of a showroom. This is an understood and widely researched fact that most men in the world need and can afford a sedan car but are actually enticed by the prospect of having a sports car. There is a deep psychological reason for this. Most men aged 40 and above who are married, for whom the car companies primarily design sedan models, have a peculiar mentality. They love the idea of having an attractive female partner with whom they can have an adventurous experience and it set them daydreaming of youth and romance but they know that these are whimsical thoughts. Car companies attract customers by showing off the convertibles on the window and when they come in, they finally choose the sedans that would be ideal to invest in. Symbolically he marries a sedan. The companies urged to put the daydream hope a little closer to the men and get a union of reality and daydream. This led to the hardtop! The hardtop convertibles were soon to become the most successful auto style introduced in the American Market for several years.

This type of insights is not a hypothesized imagination of a great business strategist but a careful and delicate analysis of the psychology of the particular customer segment.

With a huge amount of data from survey verbatim, social media, emails, transcripts etc., companies are bent over in getting such insights through open-ended text analysis. Gone are those days where the survey was just formed with close-ended questions which were quantitatively analyzed for market research. Business insights derived from customer’s open reviews, opinions give a lot of information about what they think, why they think and how they think.

Understanding the Process:

To start off the understanding of customers, there is a definitive process that one must follow through to reduce the noise in the data and to get 80% results in 20% of the effort and time.

· Define and frame the problem: More than often times, people search for solutions and move heaven and earth to understand customers opinions when they don’t even know what exactly the problem they are addressing. It is said understanding the problem is 80% of work done.

· Establish what is already known: Listing down all the facts and understanding from experience of what is known, what is believed to be known and what is expected out of the exercise to be done. Clear demarcation of facts and expectations is necessary.

· Organize data to be analyzed: It would be really ideal if the data organized and removed noise from itself. But almost every time the data has a ton of interesting insights. But the work is to streamline the data according to the problem being addressed.

· Apply systematic analysis process: This is the crux of any survey analysis and corresponding strategy to be implemented.

· Extract and create the story: This is where companies extract key insights from the data and make a wholesome report of their understanding of the problem and its solution.

Types of Open-Ended Text:

· Prompted and individual: Individual opinions and answers to specific questions.

· Unprompted and individual: Individual opinions and comments not triggered by anyone else.

· Linked with one person: Open questions to an individual with involving various matrices.

· Linked between People: People discussing products on their own in open platforms.

Ray Poynter gave a great example regarding Trip Advisor review analysis whose details can be viewed here

Common Analytical Approaches:

1. Grounded Theory: Coding the data, linking codes to concepts, linking these to categories and creating an overarching structure. Increases sensitivity to the content of the data. Inductive approach, general theories from specific observations.

2. Abductive Analysis: Identifies non-expected leaps from theories and expectations to a new one that is sufficient and probably.

3. Content Analysis: The data is coded and categorized, the frequency of codes and categories and the frequency of links between them is taken into greater account

4. Narrative Analysis: focuses on the entire text. Coding Interpreting Verifying Representing Illustrating

5. Conversation Analysis: How people speak, the pattern they use, how they create meaning. (how they say rather than what they say)

6. Thematic Analysis: Focus is to determine themes from the data.

Common Analytical elements:

· Saturated analysis: Analysis goes on till we don’t find any more new/useful things

· Structure: find/create an architecture like the above

· Make notes conclusions drawn, linking back to the data, highlighting examples

· Look to support and break Hypothesis.

Paralleldots Text APIs automates most of the organization and systematic analysis of an open-ended text.

You can read more on how to do this here and get a taste of what it does here

This post is inspired by Ray Poynter’s webinar on sense-making from an open-ended text. Ray Poynter is a celebrated Market Research Expert who is the author of the books(The Handbook of Mobile Market Research, The Handbook of Online and Social Media Research, the IPASOCIALWORKS’ The Guide to Measuring not Counting) and is currently the Founder and Chair of NewMR.

Want to improve your marketing and overall outcome. Click here to schedule a free demo.

Source: Artificial Intelligence on Medium

Leave a Reply

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

Back To Top

Display your work in a bold & confident manner. Sometimes it’s easy for your creativity to stand out from the crowd.