What is Predictive Analytics? To Predict or Not to Predict

Predictive analytics encompasses a variety of statistical techniques, including modeling, machine learning, and data mining, to analyze current and historical facts to make predictions about future or otherwise unknown events.

In business terms, predictive models exploit patterns found in historical and transactional data to identify risks and opportunities. These models capture relationships among many factors to allow the assessment of risk or potential associated with a particular set of conditions, guiding decision-making for candidate transactions.

The core of predictive analytics relies on capturing relationships between explanatory variables and the predicted variables from past occurrences and exploiting them to predict the unknown outcome. It is important to note, however, that the accuracy and usability of results will depend greatly on the level of data analysis and the quality of assumptions.

There are different types of predictive models, each of them having different purposes, and the statistical techniques underlying them vary.

Types of Predictive Models

Descriptive Models

Descriptive models quantify relationships in data in a way that is often used to classify customers or prospects into groups. Unlike predictive models that focus on predicting a single customer behavior (such as credit risk), descriptive models identify many different relationships between customers or products. Descriptive models do not rank-order customers by their likelihood of taking a particular action the way predictive models do. Instead, descriptive models can be used, for example, to categorize customers by their product preferences and life stage.

Predictive Models (or Forecasting)

Predictive models assess the likelihood that a similar unit in a different sample will exhibit specific performance. This category encompasses models used in many areas, such as marketing, where they seek out subtle data patterns to answer questions about customer performance, such as fraud detection models. Predictive models often perform calculations during live transactions to evaluate the risk or opportunity of a given customer or transaction, guiding decisions. With advancements in computing speed, individual agent modeling systems have become capable of simulating human behavior or reactions to given stimuli or scenarios.

Decision Models (or Optimization)

Decision models describe the relationship between all the elements of a decision—the known data (including results of predictive models), the decision, and the forecast results of the decision—in order to predict the results of decisions involving many variables. These models can be used in optimization, maximizing certain outcomes while minimizing others. Decision models are generally used to develop decision logic or a set of business rules that will produce the desired action for every customer or circumstance.

What are Practical Applications of Predictive Analytics?

Predictive analytics can be applied in many applications, as seen in an overview from Gartner:

Sales - CRM

Cross-Selling

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 leads to efficient cross-sales—selling additional products to current customers—which directly results in higher profitability per customer and a stronger customer relationship.

Dynamic Pricing

When it comes to pricing, predictive analytics can help analyze customers' pricing, usage, and other (buying) behavior, leading to opportunities to uplift prices without the risk of increasing customer attrition. Only uplift a price for 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.

Production Management

Failure Prediction

Techniques are designed to help determine the condition of in-service equipment, to predict when maintenance should be applied to avoid (machinery) failure. Root Cause Analysis is the starting point of building historical data to use statistics for predictive modeling.

Conclusion

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.

What Tools Can We Use for Predictive Analytics?

There are numerous solutions available which can help you with the execution of predictive analytics.

Open Source Predictive Analytic Tools

• KNIME

• R

• RapidMiner

Commercial Predictive Analytic Tools

• IBM SPSS Statistics and IBM SPSS Modeler

• KXEN Modeler (Currently picked up by SAP)

• MATLAB

• Oracle Data Mining (ODM)

• Predixion Software

• SAP

• SAS and SAS Enterprise Miner

• TIBCO Spotfire

Depending on your IT architecture and its corresponding principles, a profound decision can be made regarding the selection of the solution which serves your company best.

How to Deploy the Required Solution

Historically, using predictive analytics solutions—and understanding the results they deliver—requires advanced skills. However, modern predictive analytics solutions are no longer restricted to IT specialists. As more organizations adopt predictive analytics into their decision-making processes and integrate them into their (daily) operations, PA solutions are enabling a shift toward business users as the primary consumers.

Business users require solutions which are self-explanatory, and vendors are responding by creating new software that removes the mathematical complexity, provides user-friendly graphic interfaces, and provides intuitive out-of-the-box models that, for example, can recognize the kind of data provided to it and suggest an appropriate predictive model.

Predictive analytics solutions have become sophisticated enough to adequately present and dissect data problems, enabling any data-savvy information worker to utilize its outcome.

Lessons Leaned about Predictive Analytics

• Do not overlook the critical importance of properly, choosing, preparing, cleansing, transforming and sampling data to train and develop high performing models. (Think MONEYBALL for your business)

• Use pricipal component analysis or other attribute reduction techniques to reduce variables to avoid "over fitting" predictive models

• Be sure to partner with the business process subject matter experts to make sure all relevant aspects are captured as to not "under-fit" or incorrectly design a model

• Choose a predictive modeling algorithm that can be effectively used and deployed within a business process and not just look cool in a slide deck.

How to Use GTI in Predictive Analytics

GTI stands for Generating Transparency and Insight. It will enable your company to optimize margin and/or sales prices on a customer/product level. GTI uses predictive algorithms and applied statistics to generate business insights.

Why Use GTI?

McKinsey frequently posts articles on the power of pricing and uses statements like, "Pricing right is the fastest and most effective way to increase profits." In response, GTI takes this to the next level.

How GTI Works

Your company most probably already has a pricing tool. Here is a comparison of traditional pricing tools versus GTI:

In short, GTI uses statistics to transition from business information to business insight, making a significant impact on your pricing strategy and overall profitability.

Why choose Predictive Analytics with McCoy

At McCoy, we understand that navigating the complex world of predictive analytics can be challenging. Whether you're just starting with descriptive models or looking to optimize your decision-making processes with advanced predictive and decision models, our team of specialized consultants is here to assist you every step of the way. We offer tailored solutions that align with your business goals, ensuring that you leverage predictive analytics to its full potential.