现在，越来越多的企业重视数字化技术的持续应用，数字化运营在企业中也占据着越来越重要的位置。今天，就通过小米有品的用户增长负责人黄泱老师在大会上的演讲，为我们揭秘小米的数字化用户运营的本质及产品功能的实践。
一、数字化到底是什么？
数字化这几年挺火的，那数字化到底是什么？数字化即利用数字化技术（如大数据、云计算、人工智能等）来推动企业对内管理、对外业务上关键节点的自动化改造，从而提升整体经营效率。
其实数字化在企业当中用无非就两个点：内部的经营管理和外部。客户的经营在这两个过程当中一些关键节点的自动化的技术改造，从而提升整个经营的效果。
黄泱老师指出，数字化简单的可以概括为以下两个方面：业务的数字化和管理的数字化，围绕这两个方向，可能各个企业有些不同的数字化架构。
接下来聊聊数字化运营。运营中有一个段子：互联网公司中有三个工作—研发、产品和运营，要做研发的工作起码得学个一两年才能上岗，要做个产品可能是少得半年才能上岗，但是要做运营的工作，现在立刻马上就可以上岗。这是为什么呢？因为运营的活全都是0—1的工作，不需要任何的过渡期。
那运营应该如何找到关键节点做到标准化呢？任何事物找到目标，理清楚本质，就能找到关键节点，从而标准化，进一步就可数字化。举个例子：“财富自由的标准节点”。
假设一个八年的产品经理需要选择一个行业，这个行业的选择代表了他的成功率，接下来就是他在这个行业里头扮演的角色是什么。比如他是管理层，管理也是用杠杆，管多少人其实就是他的杠杆。如果他是总监或者VP，那么他的杠杆就是2-10倍左右，因此他的个人能力✖杠杆大概就是一年能创造的收入。如果选择创业的话，可能会更高一点，大概是100倍。
这里可以简单的计算一下，如果在30岁你的职业积累达到一定程度，到45岁之前，三年一次试错成本的话，那么还有5次财富自由的机会。这个例子只是简单的用来调侃一下，只是阐述连财富自由都有关键节点都有标准化，更不用说运营了。
二、用户运营到底是做什么的？
回到今天的主题，在聊数字化运营之前，我们先聊一聊用户运营。黄老师在京东金融做运营三年多，在互联网行业从事这么久，意识到没有人能搞明白运营到底是做什么的。为阐述清楚，需要从大方向说起，即互联网思维。那互联网思维到底是什么呢？
互联网思维不是简单的通过互联网去做一些生意，把线下的货搬到线上去卖，这只是把互联网当成了一个工具。下面列出了一些通过互联网思维做企业经营的一些标杆企业：
这些企业基本都是用户通过他们的朋友或者朋友圈突然知道的，减少了渠道之间的传播。当去研究这些企业时，会发现他们有两个共同的点：极致的产品体验和快速的口碑传播。
三、产品思维到底是啥？
黄老师认为，合格的产品经理有两个定义：首先是否是深度用户，能否引领用户，其次就是能否将自己放空，变成小白用户。比如在体验一个产品的时候，产品经理已经对产品很熟悉了，会产生惯性思维，但需要把产品推销给用户，因此需要站在用户的角度去考虑问题，以用户为中心，不要让用户思考、简约至上。通常可以在产品原型做出来后，找身边跟产品不相关的人，让他体验产品，看在没有指导的情况下他能否顺利操作。
运营最终是为产品服务的，通常好的产品是不需要销售和运营的。微信和苹果就是很好的例子。
四、数字化运营是如何做的？
这里需要提到一个行业通用的代表企业经营相对健康的公式，它是最终来自于每一个用户的「获取成本」及「终身贡献价值」，即LTV（生命周期内用户产生的价值）➗CAC（获客成本）需要大于等于3。获客成本通常都是固定的，但是生命周期的价值每个企业可能不太一样。
那么小米有品是如何拉新的呢？第一，中高端机型预装用户促活；第二，米系APP用户转化。小米本身就有自己的流量，那小米有品肯定需要去“蹭”这个流量。小米手机里有预装的app，小米游戏、小米电脑、等，那就需要将这些用户转化为小米有品的用户并且促活，并且需要制定相应的策略并执行。
企业做运营过程当中，策略是动态的、不可控的，需要根据运营关键点制定并调整策略。因此在日常运营过程当中三个关键的节点：策略、执行和数据。这也就意味着，企业如果实现三个节点的提升，那么数字化运营也就提升了。
策略即解决策略方面的问题和内容与产品动态化；执行即解决执行效率的问题和抽象运营需求标准化；数据即解决效果方面的问题，所有的结论都来自于数据。
五、产品架构该怎么进行？
运营从1.0时代起步，即纯人工运营—技术研发搭建活动，运营手动配置，奖励后返，数据SQL现跑；发展到2.0时代，即产品驱动运营—活动搭建工具，用户触达系统，数据看板等基础运营产品系统完备；到现在的3.0时代，即数字化运营—机器学习判断运营，获客自动出价，流量分发千人千面，用户分层运营节点自动触达等。3.0和2.0最大的区别就是引用了智能运营的概念，即通过机器大脑将各个策略和执行进行串联，从而形成自动化的运营。
数字运营关键节点的产品功能包括：策略功能、搭建功能和数据功能。
运营过程中需要保证策略是动态化的，要针对不同用户展示定项内容，并根据效果不断进行机器学习，将策略智能化。通过机器学习可以实现策略在后台进行实时编辑，机器算法会自动给用户匹配不同的权益，并得出最优方案，那么以前需要一个月搭建的策略，现在一天就可以完成。
策略大脑有八个体系：任务策略编排、资源触达体系、分层用户洞察、实时效果分析、联邦建模能力、多数据集合、机器学习能力、自动化A/B测试。企业的运营策略具有策略大脑的功能后，可以实现运营的数字化。营销搭建系统是指从活动搭建、素材制作、页面生成、用户触达等多个节点提供智能工具。数据需要可实时查看，需要达到关键指标的实时可视化追踪、数据间相关性的洞察与分析、智能营销建议，动态优化效果。
数字化运营带来的最大的变化是从流量体系评估策略到用户视角评估策略。企业的组织架构也会带来相应的改变，从团队由组织运营、内容运营、渠道运营等组成，到由量化分析师和算法分析师组成。
六、小米数字化运营的应用实践
目前互联网平台的用户基本上已经到了饱和的状态，因此需要做存量时代的精心化运营，即如何精细化用户，实现价值的提升。现在小米有品实施的精细化策略中涉及到：所见即所得的流程设计器；
自由组合策略、人、时间、渠道、场 景、动作；最快24小时内可策略验证结果。前端触达的提效是基于用户洞察的实时资源、素材联动。
那数字化运营如何在产品体验上进行提升呢？首先我们需要了解如何测量产品体验，这涉及到定量和定性两种方法。定量通常通过监控一些漏斗，比如查看黄金流程的用户的监控漏斗，从新用户下载、激活、注册、下单到复购以及退单，这些黄金流程会影响到经营状态。定性通常通过抽样的方法，询问用户的产品体验以及是否愿意推荐给朋友，并通过用户的反馈进行产品优化。
数字化运营最终解决的是企业的经济效益问题，企业的数字化运营最关键的问题就是数字化策略的智能大脑化。运营最大的问题是策略的执行和最终结果的反馈，数字化运营最终是解决关键节点的效率问题。
翻译：
Nowadays, more and more enterprises attach importance to the continuous application of digital technology, and digital operation occupies an increasingly important position in enterprises. Today, Yang Huang, head of user growth of Xiaomi Youpin, gave a speech at the 2021 Global Product Managers Conference to reveal the nature of Xiaomi’s digital user operation and the practice of its product functions.
1. Digitization
1) What exactly is digital?
Digital has been hot for a few years, so what exactly is digital? Digitization refers to the use of digital technologies (such as big data, cloud computing, artificial intelligence, etc.) to promote the automatic transformation of internal management and external business key nodes, so as to improve the overall operating efficiency.
In fact, there are only two uses of digitalization in the enterprise: internal management and external. The automated technological transformation of some key nodes in the two processes of customer management improves the effect of the whole operation.
Yang Wong pointed out that digitization can be summarized into the following two aspects: digitization of business and digitization of management. Around these two directions, enterprises may have different digitization structures.
2) How do operations find key nodes to standardize?
Let’s talk about digital operations. There is a joke in operation: there are three jobs in an Internet company — research and development, product and operation. To do research and development work, you need to learn at least one or two years before you can start. To do product work, it may take less than half a year before you can start. Why is that? Because operations are all 0-1 jobs, there is no need for any transition period.
So how do operations find key nodes to standardize? If anything finds its purpose and clarifies its essence, it can find its key nodes, which can be standardized and further digitized. One example: “The standard node of wealth freedom.”
Suppose an eight-year product manager needs to choose an industry that represents his success rate, and then what role he plays in that industry. For example, he is the management, management is also using leverage, the number of people is actually his leverage. If he is a director or VP, then his lever is 2 – about 10 times, so his personal ability ✖ lever is roughly a year to create income. If you choose to start a business, it may be a little higher, about 100 times.
Here’s a simple calculation: if you reach a certain level of career accumulation by age 30, the trial-and-error cost every three years until age 45, then you have five more opportunities for financial freedom. This example is simply meant to be a joke, to illustrate that even the freedom of wealth has its key nodes standardized, let alone operational.
2. User operation
Back to today’s topic, before we talk about digital operations, let’s talk about user operations. Mr. Huang has been working in the operation of Jingdong Finance for more than three years. After working in the Internet industry for so long, he realized that no one could figure out what the operation was. To elaborate clearly, need to start from the general direction, that is, the Internet thinking. So what exactly is Internet thinking?
Internet thinking is not simply through the Internet to do some business, the offline goods to sell on the line, this is just the Internet as a tool. The following is a list of some benchmarking enterprises that do business through the Internet thinking:
These enterprises are basically suddenly known by users through their friends or circle of friends, reducing the spread between channels. When we study these enterprises, we will find that they have two points in common: extreme product experience and rapid word-of-mouth spread.
3. Product thinking
Mr. Huang believes that a qualified product manager has two definitions: first, whether it is a deep user, whether it can lead users, and then whether it can empty itself and become a small white user. For example, when experiencing a product, the product manager has been very familiar with the product, will produce inertia thinking, but need to sell the product to users, so need to stand in the user’s point of view to consider the problem, user-centered, do not let users think, simple first. Usually after the prototype is made, you can find someone who is not related to the product and let him experience the product to see if he can operate it smoothly without guidance.
Operations ultimately serve the product, and usually good products don’t need to be sold or run. Wechat and Apple are good examples.
4. Digital operation
It is necessary to mention a general formula in the industry that represents the relative health of business operation. It is the “acquisition cost” and “lifetime contribution value” from each user, that is, LTV (the value generated by users in the life cycle) ➗CAC (acquisition cost) should be greater than or equal to 3. Acquisition costs are usually fixed, but lifecycle values can vary from business to business.
So Xiaomi Youpin is how to pull new? First, mid – and high-end models preinstalled users to promote life; Second, Mi APP user transformation. Millet itself has its own flow, that millet product must need to “rub” the flow. Xiaomi mobile phones have pre-installed apps, Xiaomi games, Xiaomi computers, etc., so it is necessary to convert these users into Xiaomi users and promote them, and corresponding strategies need to be formulated and implemented.
During the operation of an enterprise, the policies are dynamic and uncontrollable. Therefore, you need to formulate and adjust the policies based on the key points of operation. So there are three key nodes in the daily operations process: policy, execution, and data. This also means that if the enterprise achieves three node improvement, then the digital operation will also improve.
Strategy is to solve the problem of strategy and dynamic content and product; Execution is to solve the problem of execution efficiency and standardization of abstract operation requirements; Data is the solution to the problem of effectiveness, all conclusions come from the data.
5. Product architecture
Operation started from 1.0 era, that is, pure manual operation — technology development and construction activities, manual operation configuration, reward return, data SQL running; In the 2.0 era, the system of product-driven operation — activity building tools, user access system, data Kanban and other basic operation products is complete; To the present 3.0 era, that is, digital operation — machine learning to judge the operation, automatic bid by customers, traffic distribution of thousands of thousands of faces, automatic access to hierarchical user operation nodes, etc. The biggest difference between 3.0 and 2.0 is the introduction of the concept of intelligent operations, which is to automate operations by connecting strategies and execution through a machine brain.
The product functions of key nodes of digital operation include: strategy function, building function and data function.
In the operation process, it is necessary to ensure that the strategy is dynamic, display the content of fixed items for different users, and carry out machine learning according to the effect constantly to intellectualize the strategy. Through machine learning, real-time editing of policies can be realized in the background. The machine algorithm will automatically match different rights and interests to users and come up with the optimal scheme. In this way, the strategy that used to take one month to build can now be completed in one day.
The strategic brain has eight systems
The strategy brain has eight systems: task strategy orchestration, resource access system, hierarchical user insight, real-time effect analysis, federated modeling capabilities, multi-data sets, machine learning capabilities, automated A/B testing. After the operation strategy of an enterprise has the function of strategy brain, it can realize the digitalization of operation. Marketing construction system refers to providing intelligent tools from activity construction, material production, page generation, user access and other nodes. Data needs to be viewable in real time, real-time visual tracking of key metrics, insights and analysis of correlations between data, intelligent marketing recommendations, and dynamic optimization.
The biggest change brought about by digital operation is from the traffic system evaluation strategy to the user perspective evaluation strategy. The organizational structure of the enterprise will also bring the corresponding change, from the team composed of organization operation, content operation, channel operation, etc., to quantitative analysts and algorithm analysts.
6. Application practice of digital operation
At present, the users of the Internet platform have basically reached a state of saturation, so it is necessary to conduct careful operation in the stock era, that is, how to refine users and realize the improvement of value. Now Xiaomi Yupin implementation of refined strategy involves: what you see is what you get process designer;
Free combination of strategy, people, time, channel, scene and action. Policy verification results can be obtained within 24 hours at the earliest. The improvement of front-end touch is the real-time resource and material linkage based on user insight.
How can digital operations improve product experience? First we need to understand how to measure the product experience, which involves both quantitative and qualitative methods. Quantification is usually done by monitoring some funnel. Such as the monitoring funnel of users who view the gold process, which affects the business status of new users downloading, activating, registering, placing orders, repurchasing, and returning orders. The qualitative method is usually sampling, asking users about their product experience and whether they are willing to recommend it to friends, and optimizing the product based on users’ feedback.
Digital operation ultimately solves the problem of economic benefits of enterprises. And the most critical problem of digital operation of enterprises is the intelligent brain of digital strategy. The biggest problem in operation is the implementation of strategy and the feedback of final results. Digital operation is ultimately the effectiveness of solving key nodes.
本文由数字化转型网（www.szhzxw.cn）转载而成，来源于网络；编辑/翻译：数字化转型网默然。
免责声明: 本网站(http://www.szhzxw.cn/)内容主要来自原创、合作媒体供稿和第三方投稿，凡在本网站出现的信息，均仅供参考。本网站将尽力确保所提供信息的准确性及可靠性，但不保证有关资料的准确性及可靠性，读者在使用前请进一步核实，并对任何自主决定的行为负责。本网站对有关资料所引致的错误、不确或遗漏，概不负任何法律责任。
本网站刊载的所有内容(包括但不仅限文字、图片、LOGO、音频、视频、软件、程序等) 版权归原作者所有。任何单位或个人认为本网站中的内容可能涉嫌侵犯其知识产权或存在不实内容时，请及时通知本站，予以删除。
