Local Ad Serving via Client-Side Caching and Targeting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing advertising systems face challenges in providing advertisements with reduced latency, maintaining user privacy, and serving ads effectively in scenarios with limited or no network connectivity, as they rely on remote ad servers and may not utilize local targeting information efficiently.
Innovation Solution
Implementing a local ad serving application on user devices that stores and manages content items, including advertisements, allowing for selection and presentation of ads based on local metadata and targeting information without relying on external networks, enabling pre-fetching and serving of ads with reduced latency and maintaining user privacy by processing location data locally.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ads are served from remote ad servers, then ad delivery can be centralized and managed, but ad delivery latency increases and service availability decreases when network connectivity is limited
Solution Approach 1:
The system performs preliminary actions by pre-fetching and caching advertisement content items on the client device before they are actually requested. The ad serving application stores content items including images, videos, and associated metadata in local cache, so that when an ad request occurs, the content is already available locally, eliminating network latency and ensuring immediate delivery.
Solution Approach 2:
The patent segments the ad serving system into remote server components and local client components. The remote server handles ad selection and initial content delivery, while the local client handles caching, storage, and final ad delivery. This segmentation allows the system to benefit from both centralized management and local rapid delivery, resolving the contradiction between centralized control and fast local service.
2Adaptability or versatility
If user data is processed remotely on ad servers, then centralized ad targeting can be achieved, but user privacy is compromised and data security is reduced
Solution Approach 1:
The patent extracts sensitive user data processing from the remote server environment and moves it to the local client device. The ad serving application on the client device performs local ad selection using cached content items and device-specific information, eliminating the need to transmit sensitive user data over the network. This extraction of data processing to the client side maintains ad targeting capability while significantly reducing privacy risks.
Solution Approach 2:
The client device performs self-service by autonomously selecting and serving ads from its local cache using device-specific information and cached ad content. The ad serving application can perform local ad selection without requiring continuous communication with remote servers, enabling the system to maintain adaptability for targeted advertising while protecting user privacy through local data processing.
3Productivity
If ads are pre-fetched and cached on client devices, then ad delivery latency is reduced and service availability improves, but device storage requirements increase and network bandwidth is consumed
Solution Approach 1:
The system applies partial action by caching only a subset of ad content items on the client device rather than all possible ads. The ad serving application maintains a cache of recently requested or predicted ad content items, updating the cache selectively based on usage patterns and cache eviction policies. This approach provides fast local delivery for frequently requested ads while limiting storage consumption on the client device.
Data Source
AI summary
In general, this specification relates to content presentation. The subject matter of this specification can be embodied in, among other things, methods that include the actions of receiving a group of content items at a user device, the content items configured for presentation to a user in association with other content, each content item of the group of content items including metadata associated with the content item; identifying a request from an application of the user device to present a content item in association with particular content; selecting, using one or more processors, a content item from the group of content items using the metadata for the content items along with local targeting information; and presenting the selected content item.