Analytics Projects

Business Analytics Projects can be applied to most business functions, for example, marketing, manufacturing, customer service and operations, and risk management. Each has its own nuances, objectives and challenges.

To cater for these differences, Altis projects start with understanding the business issue, and collaboratively developing the business requirements. We then obtain source data and perform the analytics that underpins our management advice.

Following is an outline of the different data mining approaches we might employ, depending on the project.

Predictive Modelling

This helps to develop insights based on past customer behaviour. What they have bought and what actions they have taken previously.  From this we see whether they are more or less likely to buy another product or leave a business, for example.

Predictive modeling helps you focus your efforts and confine your costs. For example building marketing campaigns based on predictive lists improves Marketing ROI.

It also helps you better understand what the drivers are for certain customer behaviours. These drivers then become levers you can pull to manage the business intelligence in the business.

Text Mining

Text mining is used when the data is unstructured. For example, sales rep notes, call center notes, conversation notes, investigative notes or incident investigations. The benefits include improved customer experience through better understanding of, and responding to customer needs.

This analysis is also useful for government and insurance organisations looking to prevent fraud or optimise case management. For the legal profession it can be used to help classify new cases as they are published.

Network Analysis

Who are your most important customers? And how can they help you build your business? Network analysis can identify your customers with the greatest potential to influence other customers. This infomation can be used for a range of marketing strategies such as testing new products or promoting your business name via these important influencers.

Statistical Customer Segmentation

Finding customer usage data, rather than sales data, helps your business understand customer needs and their intention to respond. This analysis leads to the optimal number of customer segments and the behavioral criteria that are the biggest indicators of segment membership.

By looking beyond advertising segmentation, your organisation can gain a new level of confidence when measuring a customer’s value or loyalty, and their importance to your marketing strategy.

Market Basket Analysis

An effective way to identify the low hanging fruit for cross sell and up sell opportunities. Market Basket Analysis uses probabilistic methods, in combination with statistics, to find the most valuable product combinations.

For example, market basket analysis comparing your website browsing profile with other browser’s purchase behaviour, will recommend products most relevant to you. As well as increased sales this creates customer intimacy and improved retention.