Shulou(Shulou.com)11/24 Report--

Whenever a new technology is born, enterprises or individuals with forward-looking vision will take the initiative to explore application paths and scenarios. Today is obviously the age of the Big language Model (LLMs)!

Rapid growth of market demand and productivity reform brought about by innovative technology

In July 2023, KPMG released the 2023Q1 China Consumer goods Retail Quarterly report, which pointed out that the Chinese economy showed signs of consumption recovery, and retail sales rebounded significantly in the first quarter of 2023. With the continuous improvement of residents' disposable income and the active support of national policies, the industry is facing a new competitive pattern and business model.

In the same year, the wave of large language models swept the world. Erik Brynjolfsson, Danielle Li and Lindsey R. Raymond studied the phased introduction of conversation assistants based on generative AI based on the data of 5179 customer support agents (Customer Support Agents), and published a paper with the theme of "Generative AI at Work". The results show that the use of tools can increase productivity by an average of 14%. At the same time, suggestive evidence (suggestive evidence) shows that the AI model spreads the work experience of senior workers and helps new employees integrate into the experience curve.

The Trident combination of ▌ industry opportunity + innovative technology + reliable supplier consumer goods retail enterprises realize that the massive consumer groups and industry trends have brought new market opportunities, but also brought fierce competition. Effective communication and customer insight are the key to paying close attention to consumers, guiding strategies and adjusting costs. However, in order to really land the big language model in the vertical industry, we must adopt the supplier with industry Know-How and the vertical big language model.

Tianrun Rongtong's "Weiteng Big language Model platform" based on the field of customer contact has been verified by market customers in the Internet, automotive, retail, manufacturing and other industries. Provide enterprises with three major values: efficient assistance, complementary man-machine advantages, magnifying value output; efficient collaboration, knowledge-driven enterprise development, can provide customers with better products and services; user insight, insight into the business value hidden in massive data.

Next, we will take a well-known domestic dairy brand to apply the Weiteng big language model as a sample to unravel how it plays a role in the consumer goods retail business scene.

A case study on the Application of Weiteng Model in a Dairy Brand contact Center

The dairy brand has covered many domestic channels and markets with diversified and innovative products, and has been praised by a large number of consumers. Adhere to the combination of consumer demand and innovative technology, constantly explore the production, sales, after-sales and other aspects of cost lean control and overall performance improvement is the brand always adhere to the goal.

However, as market competition intensifies, it expects to solve the three core pain points through innovative technology:

The tricky problem of consumers brings great challenges to the professional level and knowledge breadth of manual seats.

Brand private shopping malls and third-party platforms accumulate tens of thousands of pieces of communication and comment data every year, so it is very difficult to clean and analyze the data.

Create omni-channel, platform-wide preference communication and preference maintenance for private domain members, but it is difficult to aggregate data and analyze member preferences.

With the rise of large model technology, the enterprise sees the possibility of solving the above problems and carries out pilot tests on a number of industry large models. Based on the consideration of technical capability, product landing, service level and so on, it finally selected Tian Runtong Weiteng large language model platform as the intelligent base of enterprise digital transformation 2.0.

The reason for the successful selection of the Weiteng large model platform, the person in charge of the brand's digital business believes that it has significant advantages in the vertical customer contact scene, and can effectively manage the massive product knowledge documents and customer data of the brand. it is an important platform for business growth and cost control.

Tian run Rong Tong mainly completes the implementation and business upgrade from four key steps.

1. Activate the new driving force of enterprise growth based on five major contacts and three steps.

Retail brands usually interact with consumers through five contact platforms: private platform: self-built mall, self-built APP, self-built Mini Program, official account, enterprise micro group, customer service hotline and other public domain media platforms: Wechat, Weibo, Douyin, Kuaishou, Xiaohongshu, Baidu and other public domain e-commerce platforms: Taobao, Tmall, JD.com, pinduoduo, Douyin, Kuaishou and other O2O platforms: ele.me, Meituan, Dianping and other offline channels: stores, Shang Chao KA channel, CVS channel, etc.

For the five major contact data, Tianrun Rongtong completed the full activation through three steps.

First of all, the daily conversation information of the first party data (First-party data) platform with private domain as the core and the open dialogue of the public domain platform, which the enterprise has the ability to operate independently and controllably, are fully connected to the Weiteng big model platform.

Secondly, for the public domain platform without open interface, the user conversation information and comments can be fully docked to the Weiteng big model platform by manual export.

Finally, the early precipitated historical conversation data, product knowledge, related knowledge, activity information, historical unidentified problems and so on are comprehensively introduced. In this way, the first step of the enterprise knowledge base is to introduce the knowledge base.

After we have done the preparatory work, we will use the extraction, aggregation, analysis and five core technology engines of the Weiteng model platform. the key information in knowledge, gold medal speech, customer common questions, poor evaluation reasons, activity description, comprehensive collation into agent-assisted robot, text robot, conversation analysis and other modules, and finally directly applied to the whole platform communication and business optimization process.

2. The combination of knowledge base and text robot can greatly reduce the conversion rate.

As consumers pay more and more attention to diet health, they pay great attention to food ingredients and other related contents when shopping. Therefore, the problems received by customer service in various channels of the brand are often very professional, which is a severe test to the knowledge reserve and service ability of customer service, and therefore often affects the service efficiency.

After connecting with the Weiteng big language model, the agent-assisted robot can recommend the best speech to the agent through context dialogue analysis and the information in the knowledge base to help it quickly complete the service process.

At the same time, because the early text robot FAQ is difficult to maintain, the recovery ability is weak, and the response corpus is stiff, the conversion rate is very high. Combining the enterprise knowledge base, intelligent error correction, context analysis, multi-round dialogue, document question and answer, automatic filling and work order automatic circulation of the Weiteng language model with the Tianrun text robot, the conversion rate is significantly reduced by 46.1%.

3. Knowledge extraction + conversation analysis + similar clustering to accurately guide brand management.

The brand produces hundreds of thousands of comments on the online platform every year, and the management of comments and the extraction of valuable information has always been its "nightmare". Although the brand has been using the AI intelligent tool of deep learning + keyword extraction technology, a large number of repetitive, similar meaning, invalid content can not be effectively cleaned.

Now, as long as the brand regularly imports the comment data of each platform into the Weiteng language model platform, it can efficiently generate multi-dimensional consumer data analysis, and the marketing team can quickly adjust the marketing strategy, pricing strategy and product taste.

4. intelligent dynamic data analysis to realize thousand people and thousands of faces 2.0.

To achieve the ultimate membership service, so that members from every purchase, each activity can get personalized experience is the goal that the brand has always wanted to achieve. After joining the Weiteng big language model, it helps the brand to maintain a large amount of user information globally in the way of OneID. Consumers can give thousands of responses to thousands of people according to their historical orders, purchase quantity, multi-platform messages and historical customer aspirations no matter which platform or activity they access. According to brand feedback, since using Weiteng app for 5 months, the praise rate of the whole platform has increased by 27.1%, and the activity participation has also increased by 31.5%.

Unlock enterprise growth: customer contact + large model + enterprise knowledge

The application of Weiteng big language model in the customer contact center of a dairy brand is a successful example of digital transformation. The value it brings is not only presented at the data level, but also reflected in the establishment of a closer, deeper and more lasting relationship between brands and consumers. Weiteng big language model platform is not a technical platform, but an important platform to drive enterprise development and business innovation.

In the future, Tianrun Rongtong Weiteng big language model will release more potential to enable the rapid growth of enterprise customers in various industries!

Follow the official Wechat account of "Tianrun Rongtong" to open a new chapter in enterprise growth.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.