image-exclusive-meghna-varma-and-misha-mehta-mediabrief.jpgMeghna Varma, Director, Marketing and Misha Mehta, Manager – Market Insights at IPM India Wholesale Trading Pvt Ltd (country affiliate of Philip Morris International, Inc.), writing exclusively for MediaBrief on invitation, outline how efficient Market Research (MR) can identify the gaps in marketing strategies, how AI has empowered researchers with insights-backed solutions, the growing applications of MR, AI in business and more.

Meghna Varma, Director, Marketing, IPM India Wholesale Trading Private Limited (a country affiliate of Philip Morris International), joined IPM India in 2005 as a management trainee and progressed through various positions in the Marketing and Sales Function over the last 16 years.

Misha Mehta, Manager – Market Insights, IPM India Wholesale Trading Private Limited (country affiliate of Philip Morris International, Inc.), works as the lighthouse to the business by helping them understand the core consumer and build potent and sustainable brands. Her expertise lies in Consumer Behaviour, Customer Insight, Fast-Moving Consumer Goods, Data Analysis, and Quantitative Research. Prior to IPM India, She was associated with brands like Reckitt and The Coca Cola Company.

One can substantially say, Market Research (MR) is the lighthouse for consumer products companies. It provides empirical evidence in data and insights for businesses to plan their portfolio and go to market strategies. Efficient MR can help organisations identify the gaps and help them develop a focused way forward. 

As the world makes leaps towards digital transformation, embracing new methodologies and resources – one important facet of technology, has been the adoption of Artificial Intelligence (AI). However, when it comes to MR, one can observe that AI has been rather under leveraged.

The advent of AI has empowered researchers with solutions backed by insights and analysis, which wouldn’t have been possible a decade back. Most notably it has the ability to process large, unstructured datasets.

When we look at examples of AI’s application in MR, we can identify two business pain points across categories. These include understanding the consumer’s emotional response to the stimuli and getting real-time reports for faster response.

AI’s application in MR is growing and the full potential is yet to be realized, yet some tools have started to show an encouraging impact on the industry – and it’s time for us to start bringing them into the mainstream and ensure better result-oriented solutions for organisations. 

Consumer’s emotional response

Daniel Kahneman in his book “Thinking fast and slow” talks about two systems of decision making, System 1 – the one we are born with (intuition/instinct/emotion); System 2 – the rational thinker.

Emotional AI tries to decode the system 1 responses. Aligning human emotional intelligence as inputs into AI, allows one to understand not only the cognitive but also the emotive channels of human thought process. This involves various inputs including facial expressions. 

Emotional AI uses video analytics to classify and quantify emotions through micro-expressions on the respondent’s face and voice and understanding the surety of the claimed response based on the secondary visual input.

It can help the business create clear differentiation among stimuli by adding a layer of emotional appeal, and in a fraction of the time that the traditional methodologies used. This can be observed in the eye gaze analysis. The interest of the respondents is very transparent. 

For such a tool, a stimulus is the most critical, subconscious reaction to every type of stimulus is different and we need to analyse results, accordingly, using the right benchmarks.

For instance, an analysis of the perception of colours in OOH advertising will help researchers identify the most preferred colours that potential buyers will be more attracted towards however if the research aims to find where on the screen observers focus the most? – that will need a different benchmarking altogether.

Real-time distribution tracking

Let’s look at tracking of the distribution process: wherein two methods are traditionally used – counting the stock and asking the retailer. Both these approaches are time and cost-intensive, thus prone to biases. Also, by the time the report reaches the stakeholders, the critical time has already been lost.

With the use of image recognition, the business can identify if the relevant Stock Keeping Unit (SKU) is present in an outlet. This can be done by an interviewer going to a retail outlet, clicking images of all the stocking points (visible or hidden) through a customized camera app.

The images then get uploaded to the servers and processed by image recognition software. Post-training, over a period, the Image Recognition (IR) AI can identify SKUs through unique markers or QR/bar codes. The presence or absence of the stock at the SKU in an image would define its availability.

That said, such a tool would come with some challenges – it’s best suited for smaller outlets and it takes considerable time and effort to train the AI through actual in-market images – so planning needs to start at least a year in advance. And there will be many more challenges as the adoption increases.

IR enables distribution reports within 3-4 days of store visit vs. the usual 15-day lag. It cuts down tracking costs by half. This tool can be used beyond distribution tracking for use cases like auditing adherence to planogram guidelines and tracking share of shelf.

As Edison said, “Value of an idea lies in using of it”, in the present day and time, technology will be the key to unearth insights of consumer behaviour when the consumption patterns are evolving at a break-neck speed. Deciphering the right trend at the right time can make or break businesses in a post-pandemic era.