Market segmentation and market movement solutions
Overview
Having a market segmentation solution in place means that users can easily and accurately define customer or client segments by chosen parameters or attributes such as geographic, demographic, behavioral, and psychographic. Based on this solution they can adapt their marketing and sales strategy, but they can also optimize other business processes in line with targeted segments. For a company, a software solution like this can lower costs and optimize resource allocations.
Our approach was to analyze data based on consumers’ behavior and their attributes to properly distinguish them and separate them into appropriate segments. Not only that, but the solution had to continuously gather data from multiple data sources to show the current status of these consumers. The first step was to analyze data sources and data structure to determine what could be used to define segments and what data will provide the best results. It was important to see what internal and external sources could help in determining consumer or client characteristics and attributes.
Our solution focused on the B2B market in the insurance industry to create precise segments so sales and marketing communication strategies could be created accordingly. This allowed for the market to be shown more accurately by outlining untapped possibilities. Based on defined segments, a predictive model was applied so future market movements can be forecasted. With this part of the solution, users can identify what changes in the consumers’ behavior and if they will shift to another segment. This minimizes uncertainty and gives the opportunity for departments inside the company to adjust their business plans and strategies.
Project scope
Discovery:
- Client interviews
- Data validation
- Project planning
Procedure:
- Data cleansing
- Data analysis
- Baseline model implementation
- Client segmentation (grouping)
- Forecasting market movement by segments
- Web-based GUI implementation
Project goal:
- The goal was to develop a market segmentation solution that will define market segments more precisely and accurately than the current approaches. The second goal was to predict segments’ future market movements
Methods:
- Machine learning methods and statistical analysis
- Clustering algorithms (Gaussian Mixture Models)
- Time series analysis (ARIMA)
Project duration:
- 2 months and ongoing
Technologies used:
Pmdarima
Statsmodels
Matplotlib and Seaborn
Machine learning library
Statistical library for time series analysis
Statistical library for statistical hypothesis testing
Visualization libraries
Web-based interactive development environment
Main goals:
Main issues:
Main goals:
- Define segments on the B2B market according to firmographics attributes and client’s conditions for segmenting
- Define clear segment lines and size
- Forecast future segment market movements
- Provide visualization of insights and metrics resulting from data
Main issues:
- Data understanding caused by anonymization
- Consolidation of multiple data sources with different types of data
- Data inconsistency
- Limited data sources
Our approaches:
A market segmentation software solution is a data-based solution that collects data from multiple sources on consumer or client attributes. It consolidates multiple data sources to identify common and joint characteristics (geographic, demographics, firmographic). Its main advantage is that it continuously gathers data through software which eliminates the need for manual inputting. It also diminishes the need to analyze each data source separately. An added feature is that it provides analytics for collected data. After defining market segments, it can predict future segment movements and behavior based on historical data.
Solution:
The main task of a market segmentation solution in the B2B insurance market was to divide the market and clients according to the company’s size. The second task included creating a predictive model of market movements by predefined parameters and the company’s size.
The objective was to recognize segments in a more precise way. The parameters that were used for segmentation were mainly revenue, assets, and the number of employees. Our process involved the elimination of irregularities and errors in data, data transformation, and data modeling. Through the process, we got to the important data that allowed us to get the best results. Market segments of B2B insurance buyers that were defined, presented the market more concisely and accurately than before. When the segments were established, a predictive model was applied to forecast future market movements, so it was clear how those insurance buyers will behave. This solution brings benefits such as clearer marketing communication, optimal sales strategy, better customer retention, and brand loyalty. Furthermore, it optimizes cost efficiency and resource management. It also allows for targeting unutilized market segments that haven’t had their needs met.
Results:
More precisely defined segments
Defined a new model and approach to market segmentation
Recognition of new segments
Future market predictions
Continuous data intake
Insights and metrics on market segment characteristics and attributes
Admired by
