Social Network Ad Conversion Tracking via Time Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Online advertising faces challenges in tracking conversion events across multiple devices, leading to inflated effectiveness metrics and overpayment by advertisers due to users performing conversions on different devices than where they were initially exposed to advertisements.
Innovation Solution
A social networking system monitors onsite and offsite conversion events, storing advertisement history and using a threshold time period to determine if a conversion event is credited to an advertisement's impression, thereby accurately charging advertisers based on genuine conversions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conversion events are tracked across multiple devices without verification, then the system can capture more potential conversion events, but the measurement precision of advertisement effectiveness is degraded due to false conversions
Solution Approach 1:
The system implements feedback by comparing conversion event data against stored advertisement impression data. When a conversion event is detected, the system retrieves the user's advertisement history and checks whether the user was previously exposed to the relevant advertisement, using this feedback to validate whether to credit the conversion to the advertisement.
Solution Approach 2:
The system introduces an intermediary verification mechanism that acts as a mediator between conversion event tracking and advertisement effectiveness measurement. This intermediary layer validates conversions by checking advertisement exposure history, ensuring that only legitimate conversions are counted toward advertisement performance metrics.
2Reliability
If all conversion events are credited to advertisements regardless of timing, then the system maximizes advertiser accountability, but advertisers overpay for conversions that would have occurred anyway
Solution Approach 1:
The system applies parameter changes by introducing a time threshold parameter that determines whether a conversion event occurs within a valid time frame after advertisement exposure. Conversions occurring beyond this threshold are not credited to the advertisement, adjusting the accountability parameter based on temporal proximity.
3Measurement precision
If the system tracks advertisement impressions and conversion events with verification, then the measurement precision of conversion attribution is improved, but the device complexity of the tracking system increases
Solution Approach 1:
The system achieves universality by using a multi-functional approach where the same infrastructure stores both advertisement impression data and conversion event data. The verification process leverages existing user profile structures and data storage mechanisms, allowing the system to perform multiple functions (tracking, storing, and verifying) without proportionally increasing complexity.
Data Source
AI summary
When the social networking system selects an advertisement for presentation to the user by a client device, a content identifier associated with the selected advertisement is stored in an advertisement history associated with the user. When the user performs a conversion event associated with the selected advertisement, the social networking system receives a content identifier and a user identifier from a client device used that performed the conversion event and retrieves the user's advertisement history. If the retrieved advertisement history includes a content identifier for the selected advertisement matching the content identifier received from the client device and a time between the conversion event and providing the selected advertisement to the client device is less than a threshold time period, a fee is determined for presentation of the selected advertisement.