"Artificial Intelligence + Big Data" drives customer service transformation and upgrading

Since its establishment in 2012, the State Grid Customer Service Center has achieved unified standards and intensive operations within the power supply service scope of State Grid Corporation of China through the centralized customer service of 27 provincial power companies. It has established lean service standards and management systems in terms of business, operation, quality, and human resources, and comprehensively promoted the use of new technologies and new means to promote the transformation of customer service to mobile, online and intelligent. Since 2016, the State Grid Customer Service Center has vigorously promoted the research of artificial intelligence and big data in the field of customer service, and built an artificial intelligence service system and a big data capability platform.

As the front-line window directly serving customers, the Southern Branch of State Grid Customer Service Center adheres to the customer-centric approach, applies the research and development results of artificial intelligence and big data to the entire service chain of front-end, mid-end and back-end, and continuously improves and upgrades it, striving to provide customers with intelligent, precise and personalized services, and promote high-quality transformation and upgrading of customer service.

Integrate into the support end and share data information

Traditional customer service mainly relies on business information systems, single knowledge bases, etc. for service support. When serving customers, customer service specialists have problems such as low support efficiency and poor friendliness. They mainly rely on the business skills of the customer service specialists themselves to answer customer requests, and there is a "ceiling" in improving service quality and efficiency. With the accelerated advancement of the ubiquitous power Internet of Things, customer-side terminal equipment data and company-side service resource information are gradually collected and shared in real time. In particular, the integration of marketing, distribution and dispatching has realized the same source, uniqueness, sharing and integration of professional data. In the process of customer service, the breadth and depth of information and data acquisition have been greatly improved.

By obtaining power-off records and relay status, the Southern Branch Center's business hall can proactively and accurately judge and handle complaints such as frequent power outages, reducing mis-dispatching and mis-dispatching; by promoting the application of intelligent knowledge bases, the search function is more powerful and faster, and the knowledge point matching is faster and more accurate. At the same time, animations and videos are introduced to present knowledge points to meet the needs of different customers in a diversified manner. The Southern Branch Center also built an intelligent service assistant and created intelligent seats, integrating and realizing intelligent functional applications such as "customer portrait push, process speech recommendation, real-time solution search, and intelligent auxiliary form filling", which significantly improved the ability, efficiency and quality of customer service. The Southern Branch Center deployed self-service application tools based on 171 basic tags and 28 holiday customer group characteristics to achieve real-time push of customer portraits, automatically identify the identity background, emotional characteristics, business concerns, and complaint tendencies of callers, flexibly adjust service strategies and plans, provide more proactive customer service, and give customers a better service experience.

Integrate into the operation side and promote the construction of intelligent services

Traditional customer service relies on the "sea of ​​people" tactic to ensure service experience. More manpower is invested in all aspects of operation to ensure answering efficiency and service quality. There are problems such as low energy efficiency, high cost, and insufficient use of personnel. In the field operation, the Southern Branch Center focuses on promoting the construction of intelligent services, building multi-round self-service answering services for customers, including intelligent voice, intelligent outbound calls, online robots and voice robot services, and gradually building a service pattern of general business robot answering and complex business manual answers, cultivating service gold medal experts, and achieving reduction in manpower, quality improvement, and efficiency increase; forming a robot trainer team, conducting deep autonomous learning, and continuously improving the accuracy of robot response, with the accuracy of robot scene response reaching 87%; using the neural network time series model, conducting deep learning of call traffic prediction, optimizing the factor prediction model, improving the accuracy of call traffic prediction, and launching the automatic scheduling tool application.

The Southern Branch has also developed an artificial intelligence patrol system that can realize intelligent monitoring and early warning of the status of personnel in the seating hall; in the field of service quality control, it fully applies intelligent voice quality inspection, and through voice-to-text technology, it realizes the transformation from random inspection mode to full-scale generalized quality inspection. At the same time, it can implement customized quality inspection plans based on sensitive keywords, greatly improving quality inspection efficiency; building an empowerment cooperation model for big data and business operations, through direct empowerment, cooperative empowerment, and autonomous empowerment, it develops and applies various types of big data typical demonstration application scenarios around the difficulties, pain points and bottlenecks in operations.

Integrate the analytical end to create a value "incubator"

After nearly seven years of construction and operation, the Southern Branch has accumulated massive marketing service data, mined the data blue ocean, made the data speak, and formed data products and services.

The Southern Branch promotes the application of a panoramic view of customer service based on data association. Through the four major functions of checking topology, checking relationships, checking data, and checking views, it supports operations management and analysts to conduct front-end data exploration and rapid analysis, effectively reducing technical difficulty, expanding analysis dimensions, and improving analysis efficiency in power supply service analysis, and better assisting the company's management decisions such as power grid construction planning, service resource optimization, and emerging business development. Carry out energy consumption view product planning and design, support four theme scenario applications of power livelihood assessment, precise investment decision support, customer energy reliability assessment, and electricity fee recovery risk, assist in policy formulation, planning and construction, empower power data, and promote data asset field applications. Based on the text data translated from service recordings, semantic analysis technology is used to automatically identify the main demands of customers, integrate work order information, and conduct customer demand theme analysis, power supply service short board positioning, and group and individual risk prevention and control applications.

The Southern Branch also conducts employee portrait research, designs customer service specialist group portraits around themes such as energy efficiency, quality, labor discipline, and personality traits, and serves employee performance evaluation, staff turnover warning, and human resource optimization. It establishes an energy efficiency funnel management model, explores the utilization loss of personnel at all levels, analyzes the utilization rate of personnel in various dimensions, and realizes the digital display and online management of energy efficiency indicators.