AI‑Driven User Experience Optimization in Kuaishou E‑commerce: Evolution, Practices, and Future Outlook

Introduction With the rapid development of technology, artificial intelligence (AI) has become a key driver of innovation across industries. In the context of user experience (UX), AI not only solves traditional problems but also introduces new interaction modes and higher satisfaction. This article explores the current UX status of Kuaishou’s B‑side e‑commerce, the company’s AI‑powered improvement path, and future prospects.

Current UX Situation Since 2018 Kuaishou has invested heavily in live‑commerce, making e‑commerce a core part of the platform. However, the UX faces challenges such as inconsistent page layouts, inefficient workflows, and performance stability issues, which affect user satisfaction and can lead to churn.

AI‑Era UX Evolution The AI era brings changes beyond personalization and real‑time interaction, emphasizing intelligent interaction, data‑driven analysis, and cross‑industry application, demanding higher UX standards to stay competitive.

UX Metrics – CPS Model Kuashou defined a CPS (Customer‑Perceived‑Score) metric covering four core indicators: Satisfaction (S), Consistency (C), Performance (P), and Stability (S). The overall score is calculated as:

Experience Score = 40% * Satisfaction + 20% * Performance + 20% * Stability + 20% * Consistency

Performance : speed, response time, resource efficiency, data processing.

Stability : ability to maintain reliable service under complex conditions.

Consistency : uniformity of design, layout, interaction, and content across pages.

Satisfaction : user‑perceived value measured through surveys and feedback.

Intelligent Inspection Platform To support rapid UX improvement, Kuaishou built a self‑developed intelligent inspection platform that automatically detects UI inconsistencies, performance bottlenecks, and stability issues. The platform integrates a large‑model (Kwaipilot) and a vertical knowledge base to provide data‑driven recommendations.

The platform workflow includes eight steps:

Input reception – collecting text, images, and logs.

Input formatting – normalizing data.

Problem classification – categorizing issues (performance, stability, consistency, etc.).

Knowledge‑base retrieval – searching internal documentation.

Large‑model analysis – using GPT‑4o for deep reasoning.

Code execution – safely running code snippets when needed.

Result formatting – presenting answers as natural language, tables, or visual comparisons.

Result output – delivering actionable reports to product and R&D teams.

Key benefits reported include over 20,000 automated inspections, detection of 3,000+ anomalies, 500+ stability alerts, and a 60% improvement in platform stability.

Future Outlook AI will continue to reshape UX design and management, offering more intelligent, personalized, and efficient interactions. Ongoing challenges such as data privacy, algorithm transparency, and explainability must be addressed to ensure sustainable development.

Conclusion By combining unified design standards, the CPS metric, and an AI‑powered inspection pipeline, Kuaishou demonstrates a comprehensive approach to elevating B‑side e‑commerce user experience.