Impression Effectiveness Modeling With Time-Location Granularity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for measuring impression effectiveness are constrained by limited data granularity and delayed analysis, which can be exacerbated by data privacy regulations, leading to irrelevant or non-targeted campaign updates.
Innovation Solution
A system and method using machine learning to analyze impression data with specified location and time granularity, enabling real-time assessment of marketing campaigns by isolating data to smaller regions and narrower time intervals, and utilizing a machine learning model to detect device activities related to impressions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data is collected from disparate sources with minimum granularity, then data privacy regulations are complied with, but analysis precision and location/time granularity are limited
Solution Approach 1:
The patent segments impression data collection and analysis into multiple granularity levels (aggregate, regional, temporal). The system divides data processing into discrete time intervals and geographic regions, allowing analysis at different levels of detail without requiring collection of all possible granular data points simultaneously. This enables precise local analysis while maintaining overall system manageability.
Solution Approach 2:
The patent adds temporal and spatial dimensions to traditional impression data analysis. By introducing time intervals and geographic region parameters, the system transforms single-dimension aggregate analysis into multi-dimensional analysis, enabling precise measurement of impression effectiveness across different locations and time periods without proportionally increasing data collection complexity.
2Reliability
If data spans weeks or years following impression campaign conclusion, then comprehensive analysis is possible, but assessment delay occurs reducing campaign relevance
Solution Approach 1:
The patent implements preliminary data structuring and organization during the impression campaign execution phase. By pre-tagging and categorizing impression data with temporal and spatial metadata as events occur, the system prepares data for rapid analysis without requiring post-campaign data aggregation. This preliminary organization enables quick retrieval and analysis of specific time-period data, reducing assessment delay while maintaining comprehensive analysis capability.
Solution Approach 2:
The patent implements periodic analysis intervals rather than continuous or end-of-campaign analysis. The system can analyze impression data at regular time intervals (daily, weekly, or custom periods), allowing ongoing assessment and campaign adjustment. This periodic approach balances comprehensive data collection with timely feedback, enabling relevance maintenance throughout the campaign duration.
3Quantity of substance
If third-party cookies are used to collect impression data, then data collection capability is maintained, but data privacy regulations are violated
Solution Approach 1:
The patent introduces an intermediary layer between data collection and analysis processes. The system uses aggregated, anonymized impression data as an intermediary that preserves analytical value while removing personally identifiable information. This intermediary data structure enables continued impression tracking and analysis without directly collecting or storing sensitive user information, thus maintaining data collection capability while complying with privacy regulations.
Solution Approach 2:
The patent changes the parameters of data collection from individual user-level tracking to aggregate impression-level measurement. By shifting from collecting detailed user behavior data to measuring impression delivery and aggregate response metrics, the system maintains sufficient data quantity for analysis while reducing privacy intrusion. The parameter change transforms the nature of collected data from personally identifiable to anonymized aggregate information.
Data Source
AI summary
Introduced herein are methods and systems for use of machine learning to measure effectiveness of an impression with specified granularity. For example, the methods and systems herein involve inputting impression data associated with an impression into a machine learning model to assess effectiveness of an impression under specific and narrow time and location parameters, thereby enabling assessments to be conducted in a more frequent and targeted manner.