Consumer Intent Detection via Multivariate Activity Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining consumer intent and tracking transitions along a purchase cycle are limited by relying on heuristic assumptions and lack precise analysis of online activity data, leading to inefficiencies in targeted marketing and advertising.
Innovation Solution
A system that performs multivariate analysis of consumer online activity, calculating breadth, depth, and recency values to identify state transitions along a purchase cycle without assuming specific parametric distributions, using a processing system to categorize web addresses and detect outliers based on these values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If heuristic assumptions are used to determine consumer intent, then the process is simple and fast, but the precision and accuracy of intent identification deteriorates
Solution Approach 1:
The patent replaces heuristic-based mechanical judgment systems with automated computational analysis systems that process online activity data. The system uses algorithms to calculate breadth, depth, and recency values from consumer browsing behavior, substituting human intuition with systematic data-driven metrics to improve precision while maintaining operational efficiency
Solution Approach 2:
The patent introduces three key parameters (breadth, depth, recency) to quantify consumer online activity, transforming qualitative heuristic assessments into measurable quantitative metrics. By changing the assessment parameters from subjective guesses to objective data points, the system achieves higher precision in consumer intent identification
2Ease of manufacture
If preset retargeting periods are used for website visitors, then the advertising process is straightforward, but the timing precision for marketing interventions deteriorates
Solution Approach 1:
The patent replaces static preset retargeting periods with dynamic timing mechanisms that adapt to individual consumer behavior patterns. The system continuously monitors online activity and adjusts the timing of marketing interventions based on real-time calculated metrics, making the advertising process responsive to actual consumer intent rather than following fixed schedules
Solution Approach 2:
The system implements feedback loops where consumer online activity is continuously monitored and analyzed, and the results feed back into adjusting marketing timing. By using recency and other temporal metrics as feedback signals, the system optimizes intervention timing based on actual consumer engagement patterns rather than predetermined schedules
3Loss of time
If conquest-based advertising assumptions are made, then targeting can begin immediately, but the accuracy of intent prediction deteriorates
Solution Approach 1:
The patent applies preliminary action by establishing tracking and analysis frameworks in advance that automatically begin processing consumer online activity data as soon as browsing behavior is detected. The system is pre-configured with breadth, depth, and recency calculation mechanisms that immediately start evaluating consumer intent, enabling fast targeting initiation with accurate real-time predictions rather than delayed heuristic assessments
4Measurement precision
If multivariate analysis of online activity is performed, then the precision of purchase cycle stage identification improves, but the computational complexity increases
Solution Approach 1:
The patent segments the complex analysis of consumer online activity into three distinct computational components: breadth calculation (variety of properties visited), depth calculation (extent of interaction), and recency calculation (timing of activity). This segmentation breaks down the multivariate analysis into manageable modules, improving precision while reducing overall computational complexity through structured decomposition
Solution Approach 2:
The patent manages computational complexity by transforming complex consumer behavior analysis into three standardized measurable parameters (breadth, depth, recency). By changing the analytical approach to focus on these specific quantifiable parameters, the system achieves high precision in purchase cycle identification while maintaining computational efficiency through parameter standardization
Data Source
AI summary
Aspects of the subject disclosure may include, for example, obtaining a session record including properties visited by a consumer during a sample period and extracting, from the session record, a select group of properties associated with a particular category of a predetermined taxonomy of various marketplaces. A breadth value of the select group of properties visited during the sample period is determined according to a number of unique properties of the select group of properties. A depth value of the properties visited during the sample period is also determined according to an extent of consumer interaction with content of properties of the select group of properties. A change point, indicating a state transition along a purchase cycle, is determined based on variations in distributions more than one of the breadth, depth, and recency values without relying on an assumption of a specific parametric family. Other embodiments are disclosed.