Edge-Cached HTML Variant A/B Testing for Revenue
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
AB testing on edge computing systems often results in short-term performance issues due to non-optimized website versions being shown to end-users, leading to potential revenue loss and inefficiencies in traffic allocation.
Innovation Solution
Implementing edge computing systems that modify HTML webpages on edge nodes to create variant versions, track performance metrics, and dynamically adjust traffic allocation based on real-time data to optimize delivery and reduce latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional AB testing methods are used where non-optimized website versions are shown to end-users during experiments, then experiment variations can be tested, but short-term performance deteriorates and revenue is lost
Solution Approach 1:
The patent pre-builds and caches multiple variant versions of webpages on edge computing nodes before experiments begin. When an experiment is launched, these pre-prepared variants are instantly deployed to users without any build or load latency, enabling immediate A/B testing while maintaining optimal performance since all variants are already compiled and cached at the edge.
Solution Approach 2:
The patent introduces an edge computing node as an intermediary between the origin server and end-users. This edge node acts as a local server that caches and serves multiple experiment variants, eliminating the need to load pages from the origin server during experiments. The edge node dynamically routes users to different variants while maintaining local optimization, thus preserving short-term performance during A/B testing.
2Speed
If AB testing is implemented on edge computing systems with pre-built cached tests, then latency is reduced and response time improves, but system complexity increases
Solution Approach 1:
The patent segments the webpage delivery system into multiple independent edge computing nodes distributed across the network. Each edge node independently caches and serves specific experiment variants for its local user base. This segmentation allows parallel processing of multiple experiments across different nodes, reducing overall system latency while distributing complexity across many simple, identical units rather than one complex centralized system.
Solution Approach 2:
The patent creates and caches multiple copies of webpage variants on edge computing nodes. Instead of dynamically generating pages during experiments, the system pre-creates identical copies of all experiment variants and stores them at the edge. This copying approach eliminates build latency during experiments while keeping each edge node's functionality simple and uniform, thereby reducing individual node complexity.
3Productivity
If dynamic traffic allocation is implemented based on real-time performance metrics, then short-term revenue is improved, but measurement and tracking complexity increases
Solution Approach 1:
The patent implements a feedback mechanism where edge computing nodes continuously monitor performance metrics (such as conversion rates, bounce rates, and revenue) for each experiment variant served to users. This real-time feedback data is automatically collected and used to dynamically adjust traffic allocation, routing more users to higher-performing variants. The feedback loop operates locally at the edge, simplifying measurement complexity while enabling revenue optimization.
Data Source
AI summary
A computer-implemented method executed by an edge computing system to optimize the delivery and performance of HTML webpages is disclosed. This method involves transmitting a request for a webpage, receiving and modifying the webpage, and selecting either the original or modified version to respond to client requests based on a probability. Performance metrics related to the chosen version are received from client devices and tracked using tracking software. The probabilities of sending different versions to clients are updated based on performance metrics via an edge-side bias module. The disclosure also includes variations such as using generative AI or a WYSIWYG interface for webpage modification, operating in a distributed edge computing system, tracking metrics like conversion rates or revenue, and discarding low-performing versions.