What is Predictive Analytics?

Summary

In this introductory session, Brad Hill outlines the fundamentals of predictive analytics, explaining its purpose, methods, and applications. Predictive analytics involves using historical data to predict future outcomes and guide proactive decision-making. By employing techniques like simulation, statistics, game theory, and data mining, organizations can better understand and anticipate customer behavior, optimize operations, and manage fraud risks. The session delves into the types of data used for analysis and the main areas where predictive analytics is applied, such as customer, operational, and fraud analytics. Brad also discusses the process of building predictive models and the significance of model accuracy, particularly in fields like medical research and public safety.

Highlights

• Predictive analytics uses historical data for predictions and proactive decisions. 📈
• It encompasses techniques like simulation and game theory. 🎮
• Models exploit patterns to assess risk and guide business goals. 🚀
• Applications include understanding customer behavior and fraud detection. 🕵️
• Model built and tested with data: accuracy varies by use. 🧪

Key Takeaways

• Predictive analytics connects data to action, predicting future events for better decision-making. 🔮
• Key techniques include simulation, statistics, game theory, and data mining. 🧠
• Predictive models help businesses understand patterns and assess risks. 📊
• Main application areas: customer analytics, operational analytics, and fraud detection. 💼
• Model accuracy is crucial, especially in fields affecting safety and health. ⚕️

Overview

Predictive Analytics is all about peering into the crystal ball of data to foresee future trends and outcomes. Brad Hill walks us through how organizations leverage this powerful tool to transition from reactive to proactive decision-making. Using techniques like simulation and game theory, companies can predict customer behavior, optimize supply chains, and even prevent fraud before it happens.

The session dives into data types, like descriptive and behavioral, laying out how they form the backbone of predictive models. By amplifying these with attitudinal and interactional data, predictions become even more precise. Imagine knowing if a student will drop out based on their social media posts or survey responses—that's the kind of insight predictive analytics seeks!

We explore its major applications: customer analytics for targeted marketing, operational analytics for efficient asset use, and fraud analytics to detect shady activities. Brad explains how models are meticulously built using historical data, split into training and testing sets for accuracy. And remember, while a perfect model is gold in safety-critical fields, even a modest improvement in non-critical areas can be a game-changer.

Chapters

• 00:00 - 00:30: Introduction and Objectives The chapter titled 'Introduction and Objectives' is introduced by Brad Hill. He outlines the session's objectives, which are to define Predictive Analytics, discuss its applications, and explain how it functions. The session begins with a definition of Predictive Analytics sourced from a research company.
• 00:30 - 01:30: Definition and Techniques Gardner Predictive Analytics focuses on linking data with actionable outcomes, providing insights into both current states and future occurrences.
• 01:30 - 03:00: Data Types and Examples The chapter delves into how data is used to address specific business goals, such as predicting customer attrition or understanding consumer demand. It discusses the role of predictive models in capturing the relationship between various factors to assess risks or potentials. By using these models, decisions can be made proactively, leading to improved outcomes.
• 03:00 - 05:00: Importance of Attitudinal and Interaction Data In this chapter, the significance of incorporating attitudinal and interaction data into predictive models is discussed. It is noted that while most models typically use descriptive and behavioral data, adding attitudinal and interactional data can substantially enhance the accuracy and performance of these models. An example provided is the prediction of whether a student would complete their studies at college or university. Descriptive data in this context includes aspects such as gender, age, marital status, residential address, whether they are an international student, part of an ethnic minority, or other relevant categories.
• 05:00 - 07:00: Applications in Banking and Other Areas The chapter discusses the application of behavioral variables in analyzing and predicting outcomes in banking and other areas. It highlights the importance of including detailed variables such as subjects enrolled, assessment marks, attendance, and progression through studies. The chapter suggests that incorporating attitudes from student surveys or social media can provide a solid foundation for analysis, enhancing the predictive capabilities in various applications.
• 07:00 - 08:30: Three Pillars of Predictive Analytics The chapter discusses the Three Pillars of Predictive Analytics. It emphasizes the importance of feedback, commentary, and interaction data in understanding a student's state of mind and potential actions. Key indicators include difficulties in coping with workload, consideration of dropping out, email correspondence regarding university logistics, and inactivity in accessing online accounts. These elements are crucial for predicting student behavior and outcomes.
• 08:30 - 10:00: Detailed Explanation on Customer Analytics The chapter titled 'Detailed Explanation on Customer Analytics' discusses the different dimensions of customer data analysis. It mentions the importance of understanding customer behavior through various analytics dimensions such as: 1. **Descriptive and Behavioral Data:** This refers to understanding who the customer is and analyzing their banking history, including their credit and debit records across all accounts. 2. **Interactional Data:** This encompasses the customer's activity on different platforms like internet banking, including login times, and call center interactions, such as inquiries about fees. 3. **Attitudinal Data:** Although not fully detailed in the provided transcript, attitudinal data involves understanding the customer's attitudes and preferences which could potentially be gathered through surveys or other customer feedback mechanisms. The chapter seems to emphasize the need for a comprehensive understanding of the customer by integrating these diverse data types to provide better service and personalized experiences.
• 10:00 - 12:00: Operational Analytics The chapter titled 'Operational Analytics' explores the use of predictive analytics to gather insights through various channels such as polls, surveys, and social media. It emphasizes understanding customer preferences and communication preferences. The chapter outlines three primary areas of predictive analytics application: customer analytics, operational analytics, and threat and fraud analytics.
• 12:00 - 15:00: Threat and Fraud Analytics The chapter titled 'Threat and Fraud Analytics' focuses on customer analytics and its importance in helping organizations to better understand their customers. It highlights the use of prediction methods to efficiently acquire new customers, enhance the value of existing ones, and retain profitable customers for extended periods. The chapter discusses how predictive analytics can determine if an individual from a mailing list is likely to become a profitable customer or respond positively to a campaign.
• 15:00 - 18:00: Example: Predicting Customer Churn The chapter titled 'Predicting Customer Churn' focuses on applying operational analytics to anticipate and understand customer behavior and their likelihood of switching to different suppliers. It highlights how analyzing customer actions can provide insights into whether they are considering alternative products or suppliers. Besides customer-focused analytics, the chapter also covers how operational analytics applies to asset management. This includes managing both physical and virtual assets, planning inventory for supply chains, assessing component purchase requirements, and supporting production facilities, all of which aid in managing an organization's physical infrastructure.
• 18:00 - 21:00: Model Building and Validation The chapter discusses the importance of efficient capital equipment management, emphasizing strategies for maximizing capital use and minimizing downtime.
• 21:00 - 25:00: Model Deployment and Scoring This chapter discusses the application of analytics in the domains of threat and fraud detection. It highlights the use of analytics to identify suspicious or anomalous activities such as fraudulent insurance claims or money laundering. This involves monitoring the environment by leveraging a wide range of data sources and detecting unusual patterns of behavior to identify potential threats or information breaches.
• 25:00 - 26:30: Conclusion and Contact Information The conclusion chapter emphasizes the importance of mitigating risks such as fraud by strategically controlling outcomes to minimize losses and enhance results. It highlights the role of customer analytics in not merely treating customers as unique individuals but in discerning and acquiring ideal customers, as not all customers contribute positively.

What is Predictive Analytics? Transcription

• 00:00 - 00:30 hi my name is Brad Hill thanks for joining me today for this introductory session titled what is Predictive Analytics so the three objectives I'll cover in this session are what is Predictive Analytics where is Predictive Analytics used and how does Predictive Analytics work so what is Predictive Analytics here's a definition of Predictive Analytics from research company
• 00:30 - 01:00 Gardner Predictive Analytics helps connect data to effective action by drawing reliable conclusions about current conditions and future events Predictive Analytics encompasses a variety of techniques such as simulation statistics Game Theory and data mining to do this analysis and make these predictions so these predictions enable organizations to use predictive models to exploit patterns found in historical
• 01:00 - 01:30 data to address a business goal you and this goal could be something like uh customer attrition or consumer demand these predictive models capture relationships among many factors to allow assessment of risk or potential associated with a particular set of conditions which allows decisions to be guided on a proactive basis which then result in better outcomes so what kind of data do you use for this type of analysis
• 01:30 - 02:00 most commonly it's descriptive and behavioral data however by adding uh interactional and attitudinal data you can experience a significant increase in the accuracy and performance of these predicted models so as an example if we were predicting whether or not a student would complete their study at say college or university descriptive data would be Fields like gender age marital status they address whether they're an international student part of an ethnic minority or something other some other
• 02:00 - 02:30 variable like that whereas the behavioral would include the subjects they're enrolled in their assessment marks for the individual subjects whether or not they're attending their lectures or their tutorials and maybe how many years of study or how far they're progressing through the the their degree so this will provide a really solid foundation and good basis for for the analysis and prediction however if you think about it if you knew their attitudes from say a student survey or from social media if
• 02:30 - 03:00 they provided some type of feedback or or commentary around that they were having difficulty coping with the the workload or if they even stated they they were considering dropping out then that would give you a huge clue as to their state of mind and their likely actions the last area is the interaction data and that could be something like email correspondence you know about uh enrollment deadlines or inquiries about deferring studies or even if there was some you know inactivity and logging into a uni web we account um that may
• 03:00 - 03:30 have been dormant since they've enrolled or if they're very active in forums or other class activities so in their banking activity I guess the descriptive and behavioral would be um who the customer is and their banking history for their credits and debits across all their accounts their interactional data would be the activity when they logged into their internet banking or maybe the call center history when they inquired about fees for closing their credit card and their attitudinal could be gain through
• 03:30 - 04:00 polls or surveys or a gain through social media channels to gain an insight into their opinions or desires or how their preferences for new products uh or offerings and and how they would like to be contacted so given a few examples on where Predictive Analytics could be used let's take a broader look at where it applies generally speaking there are three main areas or or pillars of Predictive Analytics customer analytics operational analytics and threatened fraud analytics
• 04:00 - 04:30 so the most common of these is around customer analytics enabling organizations to better understand their customers and predict what it is that they're likely to do so by acquiring customers more efficiently growing the value of these existing customers and then retaining the profitable customers for a longer period of time all these goals are assisted by prediction for example whether an individual from a mailing list is likely to be a profitable customer or whether not they're likely to respond to a campaign
• 04:30 - 05:00 are they going to be interested into in a particular product you know whether someone behaviors indicates whether they're thinking of switching to a different supplier altogether operational analytics generally revolves around assets or processes whether it be helping to manage a physical or virtual asset from planning the right physical inventory to stock in your supply chain to assessing how many components to purchase to support p uh particular production facilities it enables organizations to manage physical infrastructure and their
• 05:00 - 05:30 Capital Equipment by ensuring the allocation of people and the cash in the most efficient manner to maximize their capital for example preventing unscheduled downtime of a truck by analyzing the service history to identify when certain parts are likely to fail based on the conditions or the environments that's working in you know in combination with the intensity and the duration rather than just saying that this part needs to be serviced every six or 12 months
• 05:30 - 06:00 the third group of applications is around the area of threat and fraud analytics here analytics is used to detect suspicious or anomalous transactions like the potentially fraudulent Insurance claim or to detect money laundering activities in this case it's about monitoring your environment by including a wide variety of data sources across multiple areas detecting suspicious behaviors to identify those threats or information breaches patterns in crime
• 06:00 - 06:30 or fraud and then control the outcomes to deliver the best response to reduce exposure you reduce loss and maximize the impact of any action that uh that's taken where appropriate so looking at a little bit more detail around customer analytics it's not just about treating each customer uniquely you know and not every customer is a good customer so being able to identify and acquire the ideal customer
• 06:30 - 07:00 that is those that will be profitable throughout their entire life cycle you allowing you to put targeted acquisition efforts in and growing that customer increasing their lifetime value through personalized upsell and cross sell efforts now letting some of these customers go is going to be perfectly fine because some of them won't pay on time others are going to cost you money and retention is not about keeping every single customer rather ensuring that you retain the the most valuable customers by
• 07:00 - 07:30 identifying the indicators that lead towards defection and proactively reaching out to out to them to make sure that they can stay as well as enhancing the customer loyalty by turning those satisfied customers into brand Advocates so let the let those other other customers that aren't profitable the the bad customers go and and go to a competitor and you that increases your competitive Advantage operational analytics is particularly relevant for manufacturing or supply chain or those in a Services
• 07:30 - 08:00 industry but when you think about operations as a little differently so think about it as as people processes and assets it opens itself up to a much wider area of applications so being able to plan operations by allocating future expenditures in the most efficient manner and having the right quantity of the right product available at the right time at the the right
• 08:00 - 08:30 location then managing the day-to-day operations looking at the uh identifying areas that you could improve existing operational processes in employee productivity and Effectiveness therefore maximizing the longevity of infrastructure equipment and you and employee performance and finally analytics to protect an organization from threat and fraud monitoring your environment by including a wide variety of data across
• 08:30 - 09:00 multiple sources within organization even externally to detect suspicious behaviors and identify these threats whether they be information breaches crime fraud whatever the case may be but allowing you to then control the outcomes to deliver the best response to reduce exposure or loss and maximize the impact of any action that's taken so for example in the case of insurance claim identifying a claim as fraudulent after you've paid out may
• 09:00 - 09:30 prove difficult to recover the costs however identifying the claim has a high likelihood of being false when the claim is actually made that notification of first loss you know whether it be through a website or a call center could mean that it's handled very differently or referred to an investigator immediately before any payments made conversely this allows those claims with a high degree of legitimacy to be fast-tracked minimizing the time for processing them and therefore increasing customer
• 09:30 - 10:00 satisfaction let's move now to look at how Predictive Analytics Works let's take the example that I want to create a model that predicts the propensity of a customer to churn that is to voluntarily cancel their services and select an alternative supplier so starting with some historical data about our customers you know in this case some demographic data and some transactional you know you'll see we have uh age gender recent activity
• 10:00 - 10:30 satisfaction marital status and the the data that we're using here contains the outcome that we're going to predict so in this case that last field churn which is colored blue so the historical data contains known outcomes which is an indicator in this case of a customer who has churned where churn is equal to T and those where they haven't left where churn is equal to F so we we can build a model here to make predictions and determine
• 10:30 - 11:00 when each outcome is most likely to occur so generally we'll start by splitting the historical data into two sets the training set which we use to build the model and then the testing set or also known as the validation set in which we use to test the model so this approach allows the model to be built and compared against similar but not identical data for validation purposes to minimize overfitting the
• 11:00 - 11:30 model so the next step is to is to build the the the model using the training set so an algorithm is used and applied to the training data to construct a set of rules that predict the value of this churn attribute which is our Target variable so that could also be called the predicted value or the dependent variable so the remaining variables are then used to construct the prediction logic and those inputs are are also
• 11:30 - 12:00 called predictors or independent variables so the result of this training process is a it could be a set of rules to that uses those input variables or the predictors to generate a prediction for that Target which in this case is whether or not the customer is likely to churn and one technique may be a decision treat which is what we can see at the bottom here once this model is built it needs
• 12:00 - 12:30 to be tested against the testing or the validation set so this is done by feeding the data through the model using the same attributes as before to generate a predicted value for that Target variable which in this case is churn so that predicted value is then recorded and compared to the actual values so looking at that first case the customer turn value was true true and the predicted value that white value
• 12:30 - 13:00 right at the end there has also come up true so it's been predicted correctly the same with the second case where it was predicted that it would not churn the third case was predicted incorrectly where they didn't in fact churn and the model predicted that they wouldn't so if we take the ratio of correct predictions to actual outcomes we can compare the accuracy of this model against others I often get asked you know how accurate does a model need to be be uh to be used and I guess the answer to
• 13:00 - 13:30 that really depends but it also has a disclaimer attached to it so if we're talking about medical research or Public Safety or an area that could affect the life of someone or general general safety then your model needs to be as accurate as it possibly can be now if we're talking about an area that doesn't involve any one of those uh it just has to be better than what you're doing now for example if you're running a direct mail campaign and you need to create a list of prospects to Target and your
• 13:30 - 14:00 response rate is currently 2 and a half% and if you can create a model that's only 40% accurate but it means you can get a response rate of 5% and you can mail to much fewer people then it's helped you in two ways because it's increased your response rate and it's also decreased your campaign cost which in turn improves your Roi dramatically so the the accuracy doesn't need to be be really high it just needs to be better than what you're doing unless of course you're dealing with you know Medical research or Public Safety
• 14:00 - 14:30 areas so after we've built the predictive model we can gain some insight into what's going on by examining the model but the real value is in the deployment of model which is what we call scoring so in this example we can take some new data which may be a list of customers that are up for renewal in the ne in in the next month or next time period and we want to predict which one of those are likely to turn and go elsewhere so that the data we feed into this model has to have the same attributes that we used in the
• 14:30 - 15:00 creation of the model but the result in this case is where we don't know the outcome these are all current customers but we don't know which ones are likely to to churn or voluntarily cancel their contracts so we can take this new data feed it through that model and the result is then going to be a prediction for each customer as to whether or not they're going to churn those that are predicted to churn might then be contacted by the retention team and offered uh a a special offer
• 15:00 - 15:30 while those that are not may be follow may not even be followed up or they may get an alternative offer for an additional product for a cross sale so this scenario uses a a group of cases to be be assessed which is known as batch scoring this approach is appropriate when there's enough lead time before an event takes place such as you know next month's renewals or a direct marketing a direct mail campaign so in in the case where 's critical one record may be used at a time and in this case we're looking
• 15:30 - 16:00 at real time scoring an example here uh if someone was purchasing items online and submitting a basket and they were faced with a relevant cross- sell offer specific to the goods that they just purchased or when someone was lodging an application for credit online having a risk assessment done to determine How likely they are to default in a line this concludes today's session on predict Analytics thanks very much for viewing but please
• 16:00 - 16:30 feel free to leave a comment below or connect via Twitter or LinkedIn to hear about new sessions or to request a future topic cheers