Sentiment analysis is killing the Focus Group: the disruption of Market Research

Published April 14, 2021·Written by Lucia Manetti

wonderflow image

The rise of new technologies is changing the way we do market research. Old methods such as focus groups are about to fade away, paving the way for faster, cheaper and more accurate research tools.

Let’s see what’s going on.

The time has come for focus groups 

“Video Killed the Radio Star” was declared during the 80s. Now television is about to be killed by Twitch and YouTube while the development of game-changing technologies happens faster than ever. 

Marketing as a field is especially involved in this process because adapting slower than your competitors means failure, period.

Focus groups have been invented to gain an edge against the competition as well. 

The first market research technique was developed during the 20s by the American psychologist Daniel Starch who stopped people by the streets and asked them if they remember specific ads in newspapers and how they felt about them. Today, we would call it “sentiment analysis” and, over the course of a century, this raw method for market research evolved into the focus group as we know it.

Focus groups sounded like a great idea to know better our potential or actual customers and they worked well for decades but then something happened, something we can simply call: the internet.

The Internet, smartphones and social media give us a plethora of new opportunities and maybe they will not kill focus groups altogether, just like the television didn’t kill the radio. Radio still has its niche but would you set a marketing campaign on the radio only?  

The same applies to focus groups.

The limits of focus groups in a digital environment

The digital environment offers so many ways to catch the voice of customers in every stage of a product life-cycle. Currently, focus groups took place at every level of this cycle: during the development, before the launch and assisting the growth of the product.

Nowadays there are digital tools that can improve our performance throughout the whole process.

To have a clear picture of the situation, we can measure the improvement of new tech on market research on three dimensions:

  • the quantitative dimension
  • the speed dimension
  • the accuracy dimension

The focus group method is lacking something across all the dimensions, let’s see why.

focus groups

Faster and better: the digital way to sentiment analysis

The first two dimensions can be analyzed together being products of the same factor: the pure raw power of the internet.

Let’s imagine how HBO produced a new series during the 90s. In the development stage, they gathered a small focus group, as various as possible, to discover what people wanted to see. They shot a pilot and showed it in a pre-launch to the same group. After gathering reactions, they can choose to move forward or abandon the project. If the series comes out, they can assemble another focus group to get more feedback for the future.

See the problem here? 

A small number of people speaking for an entire nation in a delayed set of encounters.

Now think about Netflix doing the same thing today.

First of all, they have an immense collection of data on their own platform from which they can isolate precise consumer personas with distinctive habits. Without any proactive inquiry, they target one of them and start producing a series that will meet their taste. After the launch, they will record in real-time how people react, both using viewers data (when users drop the series, how many episodes they binge-watch in a row etc) and with a proper sentiment analysis on social media, listening to the real voice of customers.

On one hand, they can analyze not a sample, but the entirety of their audience, measuring real behaviours. On the other hand, they can read the spontaneous opinions that the more active customers voice on social media.

In front of this, relying only on focus groups is like riding a horse in an auto racing competition.

People lie: the accuracy problem

From a psychological standpoint, the accuracy problem is the most fascinating one.

It is well known in the market research field and in the broader sociological one: people lie in focus groups.

You may ask: what do they get from it?

They aren’t being questioned by police or something!

They are, in a way.

People feel judged when interrogated, always. They can feel judged by the researcher and by their peers as well. Consciously or unconsciously, the average person will answer to present themselves as they want to be perceived rather than as they really are.

On top of this, you should add the influence of the context: the moderators steering the conversation and the more extroverted people taking over.

Both sets of data we mentioned before can overcome this problem. Digital data coming from user experience can’t lie at all: they simply measure the real behaviour of the customer.

On the other hand, you can argue that with unstructured feedback -such as online reviews and social media comments– people could lie as well and sentiment analysis therefore as biased as traditional focus groups.

First of all, we should notice that these comments are spontaneous and not influenced by any researcher or moderator, removing part of the problem.

Still, we know people lie on social media too and this is why we need a tool to gather and process the unstructured feedback in a scientific way. 

This is why we need Wonderflow.

The whole internet is your focus group: the Wonderflow way to sentiment analysis

As we discussed in previous articles, Wonderflow is a software able to execute in-depth sentiment analysis from all the unstructured feedback we found on the web.

There is more: our top-notch A.I. can understand customer sentiment better than the customers themselves! 

As you can read in this case study, recognizing common patterns among thousands of texts, we can isolate precise sentiments and discover the deep meaning of recurring associations of words. 

It doesn’t matter if someone is lying because we don’t see the three, we see the forest.

Wonderflow allows researchers to work in large numbers with the precision of a qualitative method and taking a fraction of the time focus groups would take.

Maybe focus groups will always survive like the radio: an old-fashioned method to have face-to-face contact with our customers. 

But market research must go beyond a small selected group of people representing the population, it must analyse the population itself.


Share this post:

Other articles you might like:

wonderflow image

Interview · Sep 09, 2021

Voice of the Customer (VoC) Interview: Jorge Garcia, Founder at Vessel Insights & Strategy

In this edition of our Voice of the Customer (VoC) Interview series, our own Stefano Pecorari, Client Success Director at Wonderflow, welcomes Jorge Garcia, the founder at Vessel Insights & Strategy.   Vessel Insights & Strategy is a network-based insight, strategy, and internal communication agency. Before leading the company, Jorge was the Vice President at Estee Lauder, a renowned global cosmetic brand, for nearly five years. Specifically, he led the North American market, overseeing Consumer…

wonderflow image

Ratings & Reviews · Sep 16, 2021

4 Challenges in Aspect-Based Sentiment Analysis (ABSA) + Examples & Solutions

Wonderflow offers a text mining solution. Authors Anna Feltracco, Elisabetta Pisani, and the rest of our dedicated Linguistic team have shared how they perform Aspect-Based Sentiment Analysis to extract value from consumer feedback. It’s important to understand ABSA’s key challenges for our employees, customers/users, and those in linguistics, customer experience (CX), marketing, etc. They can gain a clearer understanding of this task and possibly draw inspiration from Wonderflow’s example and solutions.    At Wonderflow, we…

wonderflow image

Wonderflow · Sep 08, 2021

Meet Wanda: Why & How Wonderflow’s New Design System Improves Customer Experience

Wanda is Wonderflow’s newly-launched open-source design system built for products and digital experiences. Wanda enables a faster time-to-market, easier cooperation between teams, and facilitates creating consistent experiences across different channels.   Wonderflow is happy to publicly announce our new open-source design system: Wanda. In order, we will cover these main sections: Why we built Wanda The advantages of using Wanda How we use Wanda Our future plans Read on to learn more about the next…