Process: identifying new or dividing existing markets into smaller, focused segments with similar characteristics (e.g., demographics and psychographics) to tailor the marketing mix (products, prices, channels, promotions).
Problem: resource-intensive, costly, and prone to human bias, relying on limited pre-defined ideas. This prevents companies from exploiting new, high-potential market segments and leads to a loss of potential revenue and profits.
Use case: AI utilises alternative data (e.g., social media) and Machine Learning to analyse not only traditional segmentation variables but also identify hundreds of thousands of innovative new ones. It rapidly assesses mass combinations of these variables to find sizable, reachable market segments based on the unique aspects of a company’s offerings.
Case-study:
We used AI and data science techniques to identify the optimal location in Minneapolis for opening a car wash. To achieve this, data sets were collected from many different sources on various factors such as population distribution, income levels, transportation habits, environmental conditions, competition from existing car wash businesses and even bird abundance. Then, seven different types of analysis were undertaken using AI algorithms, machine learning models, data mining and visualisation techniques, revealing several potential regions for opening a car wash.
By utilising AI and data science techniques to conduct high-level market feasibility of geographical segments, recommendations were made with a higher level of accuracy and in less time than traditional methods.
Our research prior to developing the use case showed us that larger companies have been using AI for segmentation for several years. For example, Coca-Cola has used customer sentiment analysis and social media data to do mass personalisation of marketing. They also used insights from their drink dispensing machines to create new flavours and drinks with proven demand.