Next best actions with Redpoint
Introduction
Personalized marketing engagement is the key to building lasting relationships with customers and delivering value across their journeys. To achieve this, marketers need to be able to deliver the right message, through the right channel, with the right offer, and at the right time. This requires a deep understanding of customer preferences, behaviors, and needs, as well as the ability to act on this insight in real time.
However, deciding what is the best action to take for each customer and each situation is not a trivial task. It involves balancing multiple factors, such as business objectives, customer expectations, channel availability, budget constraints, and competitive pressures. Moreover, it requires a flexible and dynamic approach that can adapt to changing customer behavior and market conditions, as well as learn from previous actions and outcomes.
Different organizations may adopt different approaches to determine the next best actions for their marketing campaigns. Some may rely on explicit hierarchical business rules that define the logic and criteria for selecting the actions based on predefined segments and scenarios. Others may use analytical models that leverage data and algorithms to predict the optimal actions based on customer propensity, value, and risk. Some may combine both approaches and use rules to govern and refine the results of the models.
Regardless of the approach, marketers need a platform that can enable them to design, execute, and optimize their next best actions in a scalable and efficient way. Redpoint is such a platform. Redpoint provides a comprehensive solution that supports various use cases and approaches for delivering next best actions in support of personalized marketing engagement. Redpoint enables marketers to:
Manage and unify customer data from multiple sources and create a single view of the customer.
Segment and profile customers based on their attributes, behaviors, and interests.
Define and implement business rules and analytical models for determining the next best actions.
Execute and orchestrate cross-channel campaigns that deliver the next best actions in real time.
Monitor and measure the performance and impact of the next best actions and campaigns.
Learn and optimize the next best actions and campaigns based on feedback and results.
In this white paper, we will discuss how Redpoint's platform can help marketers achieve these capabilities and deliver the next best actions that drive personalized marketing engagement and customer loyalty.
What is a next best action?
A next best action is the optimal decision that a marketer can make at any given moment to engage with a customer or prospect. A next best action can be defined as the most relevant, timely, and personalized point of engagement that maximizes the customer's value and satisfaction. However, the definition of a next best action may vary depending on the marketer's or the organization's goals and strategies. A next best action can also refer to the optimal channel, message, or offer that is delivered as part of the engagement, or any combination of these aspects.
For example…
A next best action for a retailer might be to send an email with a special discount to a customer who has abandoned a shopping cart, or to show a pop-up with a cross-sell recommendation to a customer who has completed a purchase.
In healthcare, a next best action might be to remind a patient to refill their medication or schedule a follow-up appointment.
A next best action for a bank might be to call a customer who has applied for a loan, or to send a push notification with a personalized financial tip to a customer who has logged into the mobile app.
A next best action for a media company might be to suggest a relevant article or video to a reader or viewer, or to offer a free trial or a subscription upgrade to a user who has consumed a certain amount of content.
However, determining the next best action for each customer or prospect is not a simple task. It requires a deep understanding of the customer's behavior, preferences, needs, and context, as well as the business objectives and constraints. Moreover, the next best action will change over time, as the customer interacts with the brand across multiple touchpoints and channels. Therefore, marketers need a dynamic and adaptive solution that can continuously learn from data and feedback, and deliver the next best actions in real time and at scale.
Redpoint’s platform
This is where Redpoint's platform comes in. Redpoint's platform is a comprehensive solution that enables marketers to design, execute, and optimize next best actions and campaigns across the customer journey. Redpoint's platform leverages Artificial Intelligence (AI) and Machine Learning (ML) to analyze customer data from various sources, generate insights and recommendations, and orchestrate personalized and consistent interactions across channels. Redpoint's platform also allows marketers to configure their solution to suit their specific needs and goals, whether they want to optimize the next best point of engagement, the next best channel, the next best message, the next best offer, or any combination of these aspects. With Redpoint's platform, marketers can deliver the next best actions that drive personalized marketing engagement and customer loyalty.
To support next best action use cases, Redpoint offers a suite of applications and capabilities that enable marketers to manage, analyze, and optimize customer data and interactions. These include:
Redpoint Interaction (RPI): RPI is an application that allows marketers to create and manage audience segments, smart assets, and interactions across channels. Audience segments are dynamic snapshots of customers based on their attributes, behaviors, and preferences. Smart assets are reusable content elements that can be personalized and optimized for each customer and channel. Interactions are rules-based or machine learning-driven actions that deliver the right message, offer, or recommendation to the right customer at the right time and place.
Redpoint Data Management (RPDM): RPDM is an application that allows marketers to integrate, transform, and enrich customer data from various sources, such as CRM, web, mobile, social, email, and offline. RPDM also provides advanced analytics and machine learning capabilities, such as propensity scoring and recommender systems, that help marketers understand customer behavior, predict customer outcomes, and generate personalized recommendations and offers.
Redpoint Realtime Decisions (RTD): RTD is an application that allows marketers to define and execute RTD rules and smart assets that deliver the next best action to each customer in any channel. RTD leverages cached attribute lists and contexts to optimize the performance and relevance of the next best action. Cached attribute lists are staged sets of data elements that can be leveraged in execution of RTD. These could include customer details, summary data, and model scores or recommended content. Contexts represent a collection of decisions associated to where someone is engaging from, such as a section of the website, mobile app, call center, or other channel.
By using these applications and capabilities, marketers can design, execute, and optimize next best actions and campaigns that drive personalized marketing engagement and customer loyalty. Redpoint's platform helps marketers deliver the right message, offer, or recommendation to the right customer at the right time and place, across the customer journey.
Using business rules
One way to personalize marketing messages and offers for customers is to use a set of business rules executed in a hierarchy. Business rules are logical statements that define the criteria and conditions for a specific action or outcome. A hierarchy is a way of organizing the business rules in a tree-like structure, where each level of the tree represents a different level of segmentation or personalization. The following paragraphs describe how someone might use a set of business rules executed in a hierarchy to assign individuals to a segment representing their stage of the customer journey or what message might be most appropriate for them.
Example business rules might be:
Is the customer a new visitor or an existing customer?
If the customer is a new visitor, assign them to the "Prospect" segment and show them a welcome message or a sign-up offer.
If the customer is an existing customer, go to the next level of the decision tree.
Has the customer made a purchase in the last 30 days?
If yes, assign them to the "Active" segment and show them a cross-sell or upsell message or offer based on their purchase history and preferences.
If no, go to the next level of the decision tree.
Has the customer made a purchase in the last 90 days?
If yes, assign them to the "Lapsed" segment and show them a re-engagement or loyalty message or offer based on their purchase history and preferences.
If no, assign them to the "Inactive" segment and show them a win-back or retention message or offer based on their purchase history and preferences.
This decision tree can be further refined and expanded by adding more levels, criteria, and conditions, such as customer lifetime value, product affinity, channel preference, recency, frequency, and monetary value. The business rules can also be weighted and prioritized to account for different scenarios and goals. For example, a customer who is in the "Active" segment but has a low lifetime value might receive a different message or offer than a customer who is in the "Active" segment with a high lifetime value.
By using business rules in a hierarchy, marketers can segment customers based on their behavior, needs, and value, and deliver personalized next best actions that increase conversion, retention, and loyalty. Redpoint's platform enables marketers to create, manage, and execute business rules in a user-friendly interface, and to test and optimize the effectiveness of the rules using advanced analytics and machine learning. Redpoint's platform also allows marketers to leverage cached attribute lists and contexts to ensure that the next best action is relevant, timely, and consistent across channels.
This hierarchical approach is implemented as an audience within Redpoint Interaction. The audience supports the execution of complex selection, suppression, and segmentation criteria in a top-down cadence, thereby enforcing prioritization of the logic. When executed, the audience assigns each targeted individual to one (or many) segments for which they qualify, in the process tagging their record with appropriate segment metadata representing the offer/message/campaign for which they qualify. This audience can be executed as an Audience Snapshot, which re-evaluates each individual’s membership in a segment at a pre-determined interval (e.g., daily, hourly) or on-demand when triggered by an external process.
Using analytical models
In addition to business rules, analytical models can also be used to drive next best action by predicting customer behavior, preferences, and needs. Analytical models can be based on various types of data, such as transactional, demographic, behavioral, or contextual, and can use different techniques, such as propensity scores or recommendations. Propensity scores estimate the likelihood of a customer to take a certain action, such as buying a product, renewing a subscription, or churning. Recommendations suggest relevant products or services to a customer based on their past purchases, browsing history, or similarity to other customers. Both propensity scores and recommendations can help marketers identify the best offer, message, or channel for each customer and optimize the customer journey.
Redpoint's platform supports the creation, maintenance, and execution of analytical models within its own environment, as well as the integration of external models from third-party applications. This bring-your-own-model (BYOM) approach provides maximum flexibility for organizations to leverage the models and tools already in place. External models can be leveraged through the ingestion of externally-generated scores, or by making API calls to the customer’s modelling environment. Redpoint's platform provides a drag-and-drop interface for building and deploying predictive models using various algorithms, such as logistic regression, decision trees, or neural networks. Redpoint's platform also allows marketers to access models via REST APIs. Redpoint's platform enables marketers to use analytical models as part of campaign or audience execution, and to apply filters, exclusions, and prioritization rules to refine the next best action. By using analytical models in conjunction with business rules, marketers can leverage the power of data and analytics to deliver personalized and effective next best actions that enhance customer experience and loyalty.
Delivering next best actions
Delivering next best actions in real-time is a key challenge for marketers, as it can increase customer engagement and loyalty. Customers expect to see personalized and relevant offers or recommendations that match their current intent and context, whether they are browsing a website, using a mobile app, contacting a call center, or visiting a store. To meet these expectations, marketers need a way to access and deliver next best actions in real-time, across various touchpoints and channels.
One way to achieve this is to use Redpoint's RTD (Realtime Decisions) services, which are microservices that can be invoked via REST APIs. RTD services can perform various tasks, such as retrieving staged next best actions or recommendations, executing rules or models, applying filters or exclusions, and returning responses in JSON format. RTD services can be configured and leveraged according to the business requirements and use cases, and can be scaled and deployed across different environments.
Using RTD services, marketers can choose between two approaches to deliver next best actions in real-time. One approach is to use the next best actions that have been pre-computed and stored in the Redpoint CDP database and exposed to RTD as a cached attribute list. Another approach is to dynamically generate the next best actions based on the latest data and logic, which may involve integrating with external modeling services, such as AWS SageMaker or Google Cloud AI Platform, to access and execute analytical models that provide propensity scores or recommendations on the fly. Both approaches can help marketers deliver the most relevant and optimal next best action for each customer, in real-time, and across any channel or touchpoint.
Organizational strategy
Before implementing next best actions in real-time, marketers need to define an organizational strategy and pattern that aligns with their business objectives and customer needs. This involves making several decisions and trade-offs, such as:
What level of granularity and personalization is needed for the next best actions? Next best actions can be defined at various levels, such as segments, personas, individual customers, or even micro-moments. The level of granularity determines how tailored and relevant the next best actions are for each customer, as well as how complex and costly the implementation is.
What channels and touchpoints are used to deliver the next best actions? Next best actions can be delivered across multiple channels and touchpoints, such as email, SMS, web, app, call center, or store. The choice of channels and touchpoints depends on the customer preferences, the channel effectiveness, and the channel availability.
What cadence and frequency are optimal for determining the next best actions? Next best actions can be delivered in batch or real-time cadence, depending on the channel and the customer journey stage. The cadence and frequency affect the timeliness and freshness of the next best actions being delivered to execution channels, as well as the customer response and fatigue once the messages are delivered.
What content and format are most engaging and persuasive for the next best actions? Next best actions can include different types of content and format, such as offers, recommendations, messages, images, videos, or interactive elements. The content and format depend on the customer profile, the channel characteristics, and the creative design.
What approach and method are used to incorporate business rules and models into the next best actions? Next best actions can be based on business rules or models, or a combination of both, depending on the data and the logic available. Business rules can be used to define eligibility, priority, or exclusions for the next best actions, while models can be used to provide propensity scores or recommendations based on historical or real-time data.
Defining an organizational strategy and pattern for implementing next best actions can help marketers achieve a consistent and effective customer engagement across all channels and touchpoints. It can also help marketers optimize the performance and efficiency of their next best action programs by enabling them to test, measure, and refine their next best actions over time.
Conclusion
In conclusion, Redpoint provides a powerful and flexible solution for delivering next best actions in real-time, which can enhance customer experience and loyalty.
Next best actions can operate at multiple levels and mean different things to different organizations, depending on their goals and strategies.
Next best actions can be delivered in batch or real-time cadence, depending on the channel and the customer journey stage.
Next best actions can be based on business rules or models, or a combination of both, depending on the data and the logic available.
Redpoint supports a highly flexible next best action implementation model, which allows marketers to configure and leverage RTD services, as well as integrate with external modeling services, to access and deliver the most relevant and optimal next best action for each customer, in real-time, and across any touchpoint.