Applying AI – at Every Customer Interaction

Aligning Business Objectives with Dynamic Customer Journeys

Applying simple business rules to a sales strategy will enable you to identify eligible propositions for a customer. However, business rules alone will not enable you to select the ‘best’ proposition for a customer, or the proposition the customer is most likely to accept. As a result, proposition acceptance rates can be rather low when only business rules are used to make decisions. To improve acceptance rates, the business rules in a decision strategy need to be augmented with analytics. What is more, applying adaptive analytics to your strategies will enable them to detect changes in customer behavior as they occur and act on them immediately.

Modern businesses spend massive amounts of time and money in gathering data and modeling them into information that helps them to make decisions that please customers. Once built, this can work well, but till to what point?

Only till the time the customers do not change their preferences. But, will the business know when their customers are changing their preference?

What happens when the customers change their preferences?

  • Do we need to input the new data patterns/business rules to the system again?
  • Does the business at all have the data/rules for the new customer behavior?
  • What if, the customer behavior changes again?
  • Should the business be spending time and efforts on making up this information, rather than working on the real business?

Instead, the optimal customer decisioning must be dynamic, fluid and changing with the customer.

Each customer is unique in his/her own way. One size cannot fit all. Customers defer in multitude of factors, be it age, gender, geography, origin, and so on, and comes the individual in himself.

Implement Next-Best-Action Technology

Helping to determine the Next-Best-Action is the primary role of Pega Decision Management in customer-facing business operations.

Next-Best-Action is a proven method for improving the results of any type of customer interaction involving marketing, sales, service, retention, collections, risk, even data collection. Its goal is to determine the optimal action to take with a customer at a given moment under the circumstances. The optimal action is one that satisfies customer expectations while also meeting business objectives.

In a real-time inbound environment, Next-Best-Action is a process that begins with the customer taking some sort of action such as calling into the call center or responding to an email. The system then registers the customer’s intent for this action. Next, Pega Decision Management applies decision strategies to the customer’s action/intent to create a mini business case that calculates the Next-Best-Action for the interaction. Once calculated, the decision engine communicates the Next-Best-Action to the channel in which the interaction is taking place. The Next-Best-Action could be to take no action at all, or to take a service action, or to make an offer, or any number of other things.

The process begins again each time the customer responds or takes another action. This action-decision-action loop never ends for the entire lifetime of the customer relationship.

To put it in more human terms, Next-Best-Action follows a very old interaction paradigm: Listen, Learn, and Act accordingly.

To listen effectively you have to be able to remember what was said. Interaction History is what gives Pega Decision Management its long-term memory. Interaction History captures every customer response to every Next-Best-Action, even if the response is no response. Traditional business rules engines do not have an Interaction History component, so they are not capable of remembering. Pega Decision Management combines traditional business rules with predictive analytics and interaction history to determine the Next-Best-Action.

In real-time inbound environments in which an agent is involved, Pega implements Next-Best-Action via Next-Best-Action Advisor, which advises agents on how to interact with customers. Next-Best-Action Advisor is part of Pega Next-Best-Action Marketing, which applies the same principles to outbound campaigning to make all customer outreach as relevant and effective as possible.

A Business Opportunity:

Brand-name drugs vs Generic drugs.

According to FDA’s Office of Generic Drugs (OGD), Generic drugs are copies of brand-name drugs that have exactly the same dosage, intended use, effects, side effects, route of administration, risks, safety, and strength as the original drug. In other words, their pharmacological effects are exactly the same as those of their brand-name counterparts.

Customers fill prescriptions and they can be recurring. And when a brand name drug is in the prescription, the cost associated with it for the Payer as well as for the Subscriber is higher, not that the Subscriber wants it, but in most cases, the Subscriber does not know the generic drug equivalent of it.

Approach – Real-Time Decisioning & Machine Learning

A real-time decisioning engine that can propose a generic drug to a customer analyzing and understanding the uniqueness in each customer, the ever-changing customer behavior, external factors associated with the decisions, individual customer’s past intent.

Sky Solutions help organizations build a self-learning AI enabled decision engines, that grows in efficiency and accuracy along the journey.

Strategies are build that listens to customers prescriptions received across channels, and when a Brand-name drug is involved, runs an intelligent AI engine to gather the most appropriate Generic Drug that can be offered to the subscriber based on their –

  • Age
  • Health Status
  • Previous similar illness
  • Preferences over various social feeds
  • Past responses

Pega Adaptive Decision Manager (ADM) is a component that allows you to build self-learning adaptive models that continuously improve predictions for a customer. A key capability of ADM is that it can automatically detect changes in customer behavior and act on them in real-time. This enables business processes and customer interactions to be instantly adapted to a customer’s changing interests and needs.

Adaptive decisioning continuously increases the accuracy of its decisions by learning from each response to a proposition. For example, if a customer is offered a product or a drug and accepts it, the likelihood that customers with a similar profile will accept that offer also increases slightly. More precisely, the mathematical expressions of these probabilities in the model are automatically updated after each positive or negative response.

Adaptive Decision Manager is a closed-loop system that automates the model creation, deployment, and monitoring process. It can manage a large number of models without human intervention.

Source: Pega

In contrast to predictive analytics, which requires historical data and human resources to develop a reliable predictive model, adaptive decisioning can calculate who is likely to accept or reject an offer using no historical information. It does this by capturing and analyzing response data in real-time. This is particularly useful in situations where the behavior itself is volatile. A typical use case is the real-time detection of complex fraud patterns or predicting customer behavior following the introduction of a new offering.

In cases where data is available for offline modelling, predictive models can be used as an alternative, or in conjunction.

Adaptive Decisioning creates scoring models on the fly and uses them for predictions. The full adaptive modelling cycle consists of the following steps:

  • Capture response data in real time from every customer interaction.
  • Regularly:
    • Use sophisticated auto-grouping to create coarse-grained, statistically reliable numeric intervals, or sets of symbols.
    • Use predictor grouping to assess inter-correlations in the data.
    • Use predictor’s selection to establish an uncorrelated view that contains all relevant aspects to the proposition.
    • Use the resulting statistically robust adaptive scoring model for scoring customers.
  • Whenever new data is available, update the scoring model.

Adaptive decisioning can also build channel-specific models that account for differences in customer responses to outbound vs. real-time inbound offers.

When the customer or market changes, adaptive models change with them – enabling you to stay relevant and connected, always on-pace with the modern customer.

The recommendation and adaptation of Generic drug can lead to huge cost saving for both the customers and the payers and it also enables organizations to channel their efforts and money into other focus areas.