Personalized Promotion Allocation Using Intent and Budget Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing promotion systems fail to bridge the gap between customer preferences and purchasing behavior, resulting in irrelevant offers and low Return On Investment (ROI) due to non-personalized price discounts and lack of customer engagement, leading to increased churn rates and reduced profit margins.
Innovation Solution
A method and system for personalized promotion offer generation using RFM and AOV scoring, combined with a tree-based ensemble classifier and linear programming, to optimize promotion offer allocation based on customer segments, product categories, and inventory constraints, ensuring relevance and maximizing yield.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If non-personalized price discounts are offered to all customers, then the implementation is simple and quick, but the relevance to customer preferences is low resulting in poor ROI
Solution Approach 1:
The patent segments customers into distinct groups based on their purchase behavior, preferences, and engagement metrics. By dividing the customer base into segments, the system can tailor promotions to each segment's specific characteristics, thereby improving relevance without significantly increasing implementation complexity. This resolves the contradiction by enabling personalized offers through systematic customer classification.
Solution Approach 2:
The patent changes key parameters such as discount percentage, offer type, and timing based on customer segment attributes. Instead of applying a uniform discount to all customers, the system adjusts promotion parameters dynamically according to each segment's responsiveness and preferences, thereby improving ROI while maintaining manageable complexity through automated parameter adjustment.
2Loss of information
If personalized promotion offers are generated based on customer segmentation, then the relevance to customer preferences is improved, but the system complexity increases
Solution Approach 1:
The patent performs preliminary customer segmentation and preference analysis before generating promotions. By pre-processing customer data into segments and establishing preference profiles in advance, the system reduces the complexity of real-time offer generation. This preliminary action enables personalized promotions to be generated more efficiently, resolving the contradiction between personalization and system complexity.
Solution Approach 2:
The patent implements self-service mechanisms where the system automatically segments customers and generates personalized offers without requiring manual intervention. Through automated data processing and decision rules, the system serves itself in creating personalized promotions, thereby reducing operational complexity while maintaining high relevance to customer preferences.
3Ease of operation
If manual rule-based promotions are used for customer segments, then the system is easy to manage, but the personalization level is insufficient resulting in increased churn rate
Solution Approach 1:
The patent incorporates feedback loops that track customer responses to promotions and continuously refine segmentation and offer strategies. By monitoring redemption rates, purchase behavior changes, and engagement metrics, the system adjusts personalization levels dynamically. This feedback mechanism improves customer retention through better personalization while maintaining ease of management through automated adjustments rather than manual rule changes.
4Adaptability or versatility
If different promotion systems are maintained for different channels (online, store), then each channel can be optimized independently, but the overall system complexity increases and customer experience consistency is reduced
Solution Approach 1:
The patent implements a universal customer segmentation framework that works across multiple channels simultaneously. By creating channel-agnostic customer segments based on unified behavioral metrics, the system enables consistent personalization logic to be applied across online, offline, and other channels. This universal approach reduces overall system complexity while maintaining the ability to optimize for each channel's specific characteristics through parameter adjustment rather than separate system maintenance.
Data Source
Figure 1A
Figure 1B
Figure 2A
AI summary
Unlike rule based offer generation in state of the art, a method and system for personalized promotion offer generation by understanding customer buying intent to maximize return on investment is disclosed. Product and associated customer details are obtained from product catalog and transaction data. RFM score is computed based on transaction data to create customer segments capturing demographics behavioral data for customer. A tree based ensemble classifier predicts promotion offer redemption probability for each customer segment. The linear optimization model assigns an optimal number of promotion offers, to attain an increasing profit margin, thereby generating offers as per relevance. Budget constraint is applied so that maximum revenue can be attained. The final per customer allocation of the offer is based on different rules of personalization as required from the industry. The eligibility of getting gift card is computed by setting threshold of what customer should spend to get gift card.