Content Item Selection With Ghost Impression Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to accurately measure the impact of advertisements on consumer behavior, particularly in determining the effectiveness of content campaigns and conversion events, due to the lack of effective methods for distinguishing between actual and ghost impressions.
Innovation Solution
A method is introduced to identify and store ghost content impressions by matching rankings of ghost content items with actual content items, allowing for more precise measurement of content campaign effectiveness through experiments involving test and control groups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a second plurality of content items is selected based on bid values with a second quantity greater than the first quantity requested, then more content items are available for measurement purposes, but the system cannot accurately distinguish between actual impressions and ghost impressions
Solution Approach 1:
The patent segments the content items into two distinct categories: actual content items (third plurality) that are presented to users and ghost content items (first plurality) that are not presented but used for measurement. This segmentation is achieved by identifying ghost content items from the second plurality and separating them from actual content items, allowing the system to maintain a larger pool of content items for measurement while accurately distinguishing between those that receive impressions and those that do not.
Solution Approach 2:
The patent introduces ghost content items as intermediary elements that serve as placeholders or control group items. These ghost content items are inserted into the content selection process but are deliberately not presented to users, creating a controlled comparison group that enables accurate measurement of advertisement effectiveness without confounding the results with actual user exposure.
2Measurement precision
If ghost content items are identified and separated from actual content items, then measurement accuracy is improved, but the system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-identifying and tagging ghost content items before the actual content selection and presentation process. The system determines which content items will serve as ghost items in advance, assigns them special identifiers, and then uses these pre-prepared ghost items in the measurement process. This preliminary preparation simplifies the overall process by avoiding complex real-time distinctions between actual and ghost items during content delivery.
Solution Approach 2:
The patent uses ghost content items as copies or replicas of actual content items in terms of their selection process and ranking, but with the critical difference that they are not presented to users. These ghost items replicate the bidding and selection mechanism, allowing the system to maintain measurement integrity while using simpler identification and separation logic rather than complex real-time analysis.
3Quantity of substance
If the second quantity of content items is greater than the first quantity requested, then more data is available for analyzing consumer behavior, but it becomes difficult to determine actual advertisement impact
Solution Approach 1:
The patent segments the content item pool into actual presented items and ghost non-presented items, creating distinct measurement groups. This segmentation allows the system to collect data from both groups while maintaining clear information about which items actually reached users, preventing information loss and enabling accurate analysis of advertisement impact by comparing performance between actual and ghost items.
Solution Approach 2:
The patent implements a feedback mechanism where the system tracks and compares the performance metrics of actual content items against ghost content items. By continuously monitoring the differences in engagement, conversion, and other key metrics between the two groups, the system gains accurate feedback on true advertisement impact, using the ghost items as a control group that isolates the effect of actual user exposure from other variables.
Data Source
AI summary
One or more computing devices, systems, and/or methods for selecting content items for transmission to client devices are provided. A request for content may be received. A second plurality of content items associated with a plurality of rankings may be selected from a plurality of content items based upon a plurality of bid values. A quantity of content items of the second plurality of content items may be greater than a quantity of content items associated with the request for content. One or more ghost content items associated with one or more rankings of the plurality of rankings may be identified. A ghost impression indicator associated with a ghost content item of the one or more ghost content items may be stored based upon a determination that a ranking, of the one or more rankings, associated with the ghost content item matches a second ranking of a presented content item.