Install Evaluation System for Consequent Install Attribution

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing technologies face challenges in accurately tracking and evaluating installation patterns of applications, particularly in distinguishing between direct, consequent, and unrelated installs, especially when multiple application distribution activities are simultaneously active.

Innovation Solution

A method and system for improving detection of installation types of applications by using an install evaluation system that receives exposure temporal patterns, determines total, direct, and unrelated installs, and computes a consequent installs temporal pattern with increased accuracy, which is then transmitted to distribution systems to adjust application promotion activities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional installation tracking methods are used, then the system is simple to operate, but the measurement precision of install types is insufficient

Engineering Contradiction:

Improvedetection accuracy of install typesVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments installation events into distinct categories (direct installs, consequent installs, unrelated installs) and tracks each type separately using temporal pattern analysis. This segmentation enables precise measurement of different install types by analyzing exposure patterns and timing relationships between ad exposures and installation events, directly resolving the contradiction between measurement precision and system complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces temporal pattern analysis as an intermediary mechanism that bridges raw installation data and meaningful insights. By using temporal patterns as a mediator, the system can accurately distinguish between different install types without requiring complex direct attribution models, thus improving measurement precision while managing system complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If multiple application distribution activities are run simultaneously, then the productivity of promotion is increased, but the difficulty of detecting and measuring install sources increases

Engineering Contradiction:

Improvepromotion effectivenessVSAvoidinstall source attribution difficulty

Core Design Contradiction:

ProductivityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent applies segmentation by analyzing temporal patterns of each distribution activity separately and identifying which pattern precedes installation events. This allows the system to handle multiple simultaneous activities by breaking down the complex attribution problem into manageable temporal segments, enabling accurate source detection even when multiple activities run in parallel.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent utilizes periodic temporal analysis to distinguish between multiple simultaneous distribution activities. By examining the periodic patterns of ad exposures and comparing them with installation timing, the system can attribute installs to specific activities even when multiple activities are running concurrently, thus resolving the measurement difficulty while maintaining high productivity.

Inventive Principle:

Principle #19Periodic action

3Loss of information

If conventional attribution methods are used, then the device complexity is low, but the loss of information about install patterns occurs

Engineering Contradiction:

Improveinstall pattern informationVSAvoidevaluation system complexity

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The patent implements continuous temporal pattern analysis that tracks installation events throughout the evaluation period without interruption. This continuous analysis preserves complete information about install patterns, timing relationships, and exposure sequences, eliminating information loss that would occur with discrete or snapshot-based conventional methods, while managing system complexity through efficient temporal data processing.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS20250117204A1Tracking and evaluating installation patterns of applications

Publication Date: 2025.04.10 PLAYTIKA

AI summary

A method of improving detection of installation types of applications, comprising receiving an exposure temporal pattern reflecting exposure of one or more application distribution activities launched to promote installation of one or more applications, determining a total installs temporal pattern reflecting all installs of the application(s), determining a direct installs temporal pattern reflecting direct installs of the application(s) resulting from direct exposure to the application distribution activity(s), deriving an unrelated installs level reflecting installs of the application(s) unrelated to the application distribution activity(s), computing, based on the total installs temporal pattern, the direct installs temporal pattern, and the unrelated installs level, a consequent installs temporal pattern reflecting with increased accuracy consequent installs of the application(s) which consequently follow-up on the direct installs, and transmitting the consequent installs temporal pattern to distribution system(s) adapted to adjust one or more application distribution activities according to the increased accuracy consequent installs temporal pattern.