Drive business optimization with advanced analytics
The primary focus of Advanced Analytics is to apply advanced statistical models that use predictive and descriptive techniques like association, clustering, regression, correlation, decision trees, data mining, etc. in decision contexts for key business areas.
Key Features
The Advanced Analytics process is an easy to use, wizard-based, guided analytic process. It has the ability to model scenarios with multivariate statistical analysis capabilities. It can be integrated in the same interface with performance management and Business Intelligence to create a seamless user experience. It provides you the ability to optimize, save and recall model variables to deliver optimal results.
Key Applications
- Product Rationalization
- Product Affinity Analysis
- Market Basket Analysis
- Customer Churn Modeling
- Customer Segmentation
- RFM Modeling
- Sales Forecasting
Product Rationalization
Product rationalization is an intuitive, wizard-based, guided analytic process. It incorporates advanced statistical models for classification and attributes weightage control. This process helps you to improve assortment productivity and thereby optimize your profitability. It also offers flexible filters for data selection, and the ability to set degree of rationalization for generating accurate, usable deletion rules.
Product Affinity Analysis
The Product Affinity Analysis helps you to identify the items that are likely to be purchased together. The process analyzes which items are an anchor solution and drive the sales of other products. This in turn helps to determine which items you should promote to create a pull-through. This credible fact-based analysis also enhances sell-in and recommends in-store product placements. It allows you to develop optimal promotional plans that will benefit the manufacturer as well as the retailer. You can use product affinity analysis to discover products with natural affinity or even products with artificial affinity, like cigarettes and stamps.
Market Basket Analysis
Market Basket analysis provides insight into your customer profile. It helps you answer questions like who they are and why they make certain purchases. It also helps you assess which products tend to be purchased together and which are most amenable to promotion. You can gauge what merchandise should be carried and how they should be laid-out in the store. It also analyzes which products you must put on sale, when to issue coupon and to whom.
It helps you assess Basket characteristics, baskets clustering, item popularity, track marketing events and analyze product affinity.
Customer Churn Modeling
You can use Churn Management to segment customers by their value and profile by identifying the patterns of events and metrics in the customer lifecycle. It helps you to predict potential churners, establish churn alarms (with minimum false alarms) and events that allow one to adequately anticipate the point of no return. This model lets you select the Protocol of Action by churn segment and plan retention strategies. You can predict churn behavior using purchase behavior, renewal behavior, redemption behavior, demographics, etc in a regression model that scores key influencing variables.
An unsatisfied customer will share his negative experiences with at least twice as many people as he would have a positive experience. Since it is far less expensive to retain older customers than to acquire new ones, it is imperative to use the customer churn model.
Customer Segmentation
Customer segmentation is a data mining methodology useful to customize and target your marketing as well as merchandising initiatives and consequently maximize ROI. You can classify your customer base into groups of similar customers based on their behavioral and demographic attributes.
This model is used for effective marketing and promotions, merchandising planning, tracking customer's behavioral changes, new store planning and it also assists you to alter and strengthen your retail positioning.
Recency Frequency Monetary Modeling (RFM)
RFM is an effective process for marketing to your loyal customers and uses purchase behavior by recency, frequency and monetary to determine what offers work for what type of customers. Generally, only small percentages (about 5%) of customers respond to typical offers. But with RFM, you can ensure you are targeting the right set of customers who are most likely to respond. RFM is a powerful segmentation method for predicting customer response and ensures improvement in response as well as profits.
It is used primarily for targeted campaigning, customer acquisition, cross-sell, up-sell, retention, etc and is a guarantor of campaign effectiveness and optimization.
Sales Forecasting
Sales forecasting is a process based on the ARIMA time-series model for predicting the future sales of products. It provides the flexibility to create store level forecasts as well across store groups or clusters and processes the forecasts at multiple time grains (day, week, month, etc).
This process allows you to handle historic data on seasonality, promotions and new product launches to predict the future sales of your products. It creates a model based on auto correlation or partial correlation techniques and gives you the ability to optimize a model based on validation analysis.
It allows you to set realistic, accurate sales targets, forecast promotional sales lifts, and plan the demand for new products and new stores.
