Machine Learning techniques for a better Customer Experience

Machine Learning techniques for a better Customer Experience

machine-learning-customer-experience

[Figure] Consumer journey before buying (from Hugeinc.com)

One of the elements of success in online shopping, is the possibility to analyse the client, to follow them along their journey and to derive useful information that may improve their experience within the retail environment. Customer Experience (CX) Management has become a fundamental component in every marketing strategy and its attempt to increase clients’ satisfaction. With the benefit of new technologies and new analysis models, today it is now possible to provide consumer behaviour analysis for brick and mortar shops.

 

What is consumer behaviour and why is it important?

Consumer behaviour is the study of individuals or groups and their processes to select and use  products, services and experiences. The goal is to understand the needs and the impacts that these processes have on the consumer and society.

There are several factors that influence consumer behaviour, including[1]:

  • psychological factors, including motivation, perception and attitudes of consumers;
  • personal factors, such as age and life-cycle stage, occupation, economic circumstances, lifestyle and personality;
  • social factors, regarding reference groups, family, roles and status;
  • cultural aspects, concerning culture, subculture and social class system.

These factors make the buying decision process extremely complex: while many consumers move through their decision-making process in a fixed and linear sequence, for others the path can be more complex and much less predictable[2]. That’s why for marketers, trying to predict such behaviour is the key to increase consumer’s satisfaction, so that an individual’s experience meets that individual’s expectations.

 

Our Customer Behaviour Analytics tool

Konica Minolta Laboratory Europe is developing a Customer Behaviour Analytics tool that finds applications in stores, supermarkets and shopping malls, to derive information about consumers’ characteristics, satisfaction levels and their actions, and provides valuable insights to shop managers.

Our solution is based on advanced multimedia analytics technologies and it is a part of the Cognitive Hub platform. This provide a series of capabilities, including machine learning algorithms to process images for classification, semantic segmentation, environment sensing, object detection and the possibility of interacting with edge devices to perform locally heavy computational analysis. This last capability also allows the solution to remain compliant with GDPR.

The solution starts from data analytics, providing the brick and mortar stores with valuable information that they don’t have today. Such data includes:

  • gender and age of consumers, supporting sellers’ decisions on which channels should be used for marketing;
  • consumer numbers, providing a correlation between the number of customers in the shop with the number of personnel, to improve sales management and the marketing department;
  • time spent in shop, to improve the Customer Experience;
  • consumers interest detection, monitoring what they like and what they overlook;
  • customer flow inside the shop, understanding the use of space to positively influence the Customer Experience.

Through the analysis of these acquired data, the Customer Behaviour Analytics (CBA) tool provides information about:

  • consumers’ behaviour in shops, such as movement trajectories and the locations visited;
  • heatmaps of people, trolleys or other specific objects, to identify both busy and underutilized spaces, to provide crucial information for shop layout optimization, more targeted advertisement and transparent charges for space utilized by 3rd party advertisers and brands;
  • automated cashier desks open/close decision support;
  • success monitoring of temporary marketing measures – benchmarking of different locations;

This information is analysed through a statistical summarisation of the consumers data (according to the GDPR compliance) and it can then be used for improving services and for marketing, defining pricing strategy or optimising shop layouts. Furthermore, through predictive consumer behaviour modelling techniques, it is possible to predict the future behaviours of consumers, offering them the right products at the right time and in the right place.

 

Meet Konica Minolta at the Machine Learning Prague 2019

For the third consecutive year, Konica Minolta supports and participates in the Machine Learning Prague conference from 22 to 24 of February 2019. With more than 1000 expected attendees and 45 speakers, this is one of the largest events covering machine learning, artificial intelligence and deep learning applications in Europe.

The Machine Learning Prague conference can be the opportunity to meet Konica Minolta, our R&D activities and talk about how to improve your retail customer experience with our CBA tool.

Are you interested in participating? Get a special discount using the code in the image below. Come to the conference and visit the Konica Minolta booth!

machine-learning-prague-2019

 

[1] Kotler Philip, Armstrong Gary, (2008). Principles of Marketing, Pearson Education, Inc., Upper Saddle River,

New Jersey. PDF available here: http://library.aceondo.net/ebooks/Business_Management/Principles_of_Marketing(14th.Edition).pdf

[2] Ibid.

 

With contributions and edits by Giorgio Sestili.

Filip Magula
Filip Magula
Team Leader R&D Engineering
X