what is CDP? CDP (customer data platform) is a customer database managed by marketers, which collects, integrates, analyzes and applies real-time and non-real-time customer data from different channels and scenarios, so as to achieve the goals of customer modeling, designing marketing activities, improving marketing efficiency and optimizing customer experience, thus promoting the growth of enterprise performance and profit. Why do we need to build CDP contact fragmentation and multi-channel under the traditional marketing model, we are increasingly short of traffic and the traffic is becoming more and more expensive; In essence, it is not the flow that really decreases, but the fragmentation and even powdering of the contacts; Advertising has reached some target users, or customers have come to the store (online or offline), which has generated interest, even a clear intention, or has generated consumption. However, we don't know who these users are, and we can't reach them again on our own initiative. Therefore, we need to access and pull through omni-channel customer data; We need CDP as an enterprise user data hub; Extract customer data scattered in various channels and isolated systems to a unified big data platform and process the data into a unified structure, and then integrate it into a unified customer database. Refined operation With the prevalence of e-commerce, the way consumers get information has changed, and shopping habits have changed. We can't stay in the traditional way of playing, treat our users extensively, and meet the needs of consumers from toothless (babies) to toothless (elderly) in a certain area as much as possible with tearful goods. There is basically no refinement of user operations; At the same time, a large number of advertisements are wasted on inaccurate target groups, and the greater waste is that users who saw advertisements and visited stores just met us. Due to the lack of follow-up interaction, a large number of accurate potential customers have no interest, those who have interest have no intention, those who have intention have no purchase, and those who have purchase have no repurchase. Data-driven growth enterprises are not growing, then they are dying! A famous saying tells the eternal truth of "growth" in enterprise operation. Paying attention to user experience and value is the goal of enterprise growth. This core value, driven by technology and products, has led to a rapid expansion of the growth efficiency of enterprises, thus maximizing the commercial interests of enterprises in the future. Today's digital operation has gone far beyond the definition of traditional marketing, and has surpassed the concept of digital marketing in ordinary enterprises. For example, we know website marketing, WeChat marketing, search engine marketing, social media marketing and so on, and the concepts of media advertising have gradually become the necessary competitiveness of every company. Today's marketing concept has quietly extended to products, as well as sales and services and other departments and fields. It is particularly important that today's digital operation has achieved faster business growth through the research and development of core products and user experience. Especially in the retention of users, it is necessary to improve the conversion efficiency of each link, optimize in various channels at the same time, quickly find the optimized way with data, and constantly evolve and iterate. Why CDP is needed is discussed, and then how to build CDP and how to build CDP user data precipitation is discussed. In order to understand users and then carry out refined operation on users, it is necessary to collect and precipitate user data. These data include online and offline shopping behavior data of users, consumption data of consumers, interaction data with users (planting grass, drainage, transformation, fission) and so on; Only by continuously depositing and mastering three-dimensional, multi-dimensional, fresh and timely user data can we gain insight into users, know what channels users came in from, what they operated after coming in, what they watched, what products they bought, and what they jumped out of, whether it was shopping experience, product attractiveness or price; With these data, we can interact with users with different needs and preferences with high quality and frequency. The new interaction further enriches user data. After OneID omni-channel user data access and pull-through, we need a set of ideas and methods to solve the data island problem. OneID's approach is to break the data island and realize data accommodation through unified entity identification and connection. Simply put, users, equipment and other business entities will be mapped to a unique identification (UID) in the corresponding business data, and the data of each dimension will be related through this UID. The specific method is based on mobile phone number, ID card, email address, device ID and other information, combined with business rules, machine learning, graph algorithm and other algorithms, ID-Mapping is carried out, and all kinds of UIDs are mapped to a unified ID. Through this unified ID, the data of each data island can be correlated and data accommodation can be realized to ensure the accuracy and comprehensiveness of data applications such as business analysis and user portrait. Data-based operational capacity-building user data has been collected, and omni-channel user data has also been accessed through OneID. Next, we think about how to operate data. The specific approach is how to manage the user's life cycle, what marketing activities users need to take at each stage, and how to interact with users at each stage. For example, the user life cycle can be divided into acquisition, activation, retention, realization and recommendation; Through e-commerce platform, self-built micro-mall, private domain traffic operation community, mass marketing platform, and self-media matrix, various customized contents are used accurately and frequently to "plant grass" for users, thus realizing high drainage, high conversion, high joint purchase and high fission of activities. Finally, talk about what needs to be done to build CDP. What do you need to do to build CDP user tag system? Building user tag is to describe and portray the characteristics of a certain dimension of users, so that users can obtain information quickly. It can be divided into attribute tags, statistics tags, rules tags and mining tags. Through the user tag, the business personnel can quickly obtain the user's information, find the significant features, get business inspiration and realize the insight of user characteristics; Tags can enrich the dimensions of data analysis, and the data query platform can support more abundant analysis and comparison after getting through the tag data, thus increasing data analysis; User tags can cut user groups into finer granularity, supplemented by SMS, push, email, activities and other means to drive care, recovery, incentive and other strategies, so as to achieve refined operation; User tags are the basis of many data products, such as personalized recommendation system, intelligent operation platform, advertising system, CRM and so on. User life cycle management We divide the user life cycle, analyze the conversion rate of each node in the life cycle through the conversion funnel, and optimize the influencing factors of each node by combining the business objectives and operational objectives of each stage; For example, in the acquisition stage, as the product reaches the user for the first time, this is the starting point for users to use the product. At this stage, our operation goal is to attract new customers. To expand the scale of customer acquisition, we must first realize two other kinds of matching: (1) language-market matching, that is, the degree to which your description of product advantages impresses the target users; (2) Channel-product matching, that is, the effectiveness of your chosen marketing channel in promoting products to target users. -Growth Hacker therefore focuses on the matching degree between product description language and market, explores high-quality channels, and attracts users to register and use. For example, in the retention stage, the main goal is to wake up and retain users; Firstly, it is determined that different companies may have different methods to measure retention. The most basic retention indicator in the field of e-commerce is the repeated purchase rate of users, such as the number of times users buy products every month. Then screen the retained data to find out why users stay or leave. The marketing center covers six main links of digital marketing activities through the marketing center: offline marketing plan, including marketing objectives, rhythm, methods, customer base, content, personnel and implementation process. Content formulation: graphic content, coupons, courtesy, event reservation, etc. corresponding to the manufacturing scheme. Customer segmentation: accurately screen customer groups according to the purpose of the activity. For example, it is created by rules such as basic attributes, consumption indicators and membership indicators. Accurate delivery: active push (timing, cycle), triggered push (attention, registration), QR code delivery. Effect evaluation: omni-channel customer data collection; Open the registered quantity; Coupon issuing and receiving write-off events; ROI analysis and so on. Business analysis: further analyze the comprehensive effect through some business indicators, and further empower the marketing plan optimization. this text