Previously we’ve posted a blog Predictive Analysis (To predict or not to Predict) in which we’ve explained the basics of Predictive Analysis. Also the three Predictive Analytics (PA) Models (Descriptive, Predictive and Decision) are visualized and these all have to be taken sequentially in order to mature your business in Analytics.
Because this blog is about Predictive, the most probable next question (with certainty restricted probability) will be:
How can it be applied?
Predictive Analytics can be applied in many applications as to be seen in the Gartner overview below;
Sales - CRM
Corporate organizations collect and maintain abundant data (e.g.customer records, sales transactions) because exploiting hidden relationships in data can provide a competitive advantage. For an organization that offers multiple products, predictive analytics can help analyze customers' spending, usage and other behavior. This will lead to efficient cross-sales or -selling additional products to current customers, which directly leads to higher profitability per customer and a stronger customer relationship.
When it comes to pricing, predictive analytics can help analyze customers' pricing, usage and other (buying) behaviour, leading to opportunities to uplift prices without the risk to increase customer attrition; Only uplift a price at the right customer, at the right product against the right price. This directly leads to higher profitability per customer.
Inventory - SCM
S&OP Demand Planning Prediction
Using statistical Time-Series (e.g. via triple Exponential Smoothing) models to automatically calculate the “best-score” (highest R2) forecast model to automate Demand Planning, as input for your S&OP process
Techniques are designed to help determine the condition of in-service equipment, in order to predict whenever maintenance should be applied in order to avoid (machinery) failure. Root Cause Analysis is the starting point of building historical data to use statistics for predictive modelling.
After having depicted several use cases, the following conclusion can be applied:
Approach Predictive Analytics via the Business Processes axis as it’s not a technical implementation
The initiation of applied statistics goes hand in hand with a known business challenge
Predictive Analysis can best be captured via the following formula:
Blog posts are like good movies; there will always be a next one to complete the trilogy. In part III of this blog series the following questions will be answered:
What PA technology solutions are available?
How to deploy the required (solution) knowledge?
McCoy's consultants are specialized in Predictive Analytics and are more than happy to assist you in your road to Predictive Analytics
If you want to know more, feel free to contact us via our contact form