What are the three requirements for enterprise digital transformation
The planning of enterprise digital transformation needs to face several fundamental questions: what to transform? What kind of rotation is used? What kind of transformation? And these issues cannot rely solely on service providers and consulting companies to make unilateral efforts from the technology supply side. Enterprises should actively guide from the demand side, after all, digital transformation is not a simple procurement behavior of Party A/Party B, but a joint long march between enterprises and service providers.
The above three fundamental issues can be translated as: business pain points, technical anchor points, and transformation turning points. The answer is not isolated, but rather "three points and one line", which connects these three points consistently with the main line of "value creation". This article combines the author's practical cases during their time at Alibaba Cloud to analyze how traditional enterprises can play a leading role in building the "three points and one line".
Dataization of Business Pain Points - From Business Flow to Data Flow
Firstly, companies need to articulate their business pain points and explain them thoroughly.
Almost all business departments within an enterprise have pain points and itch points. Simply listing their respective challenges is just the collection of raw information, which needs to be analyzed, summarized, and sorted out. Especially, it is necessary to identify which pain points have generated serial reactions in the business process and are transmitted to the final business indicators. Sort out the isolated pain points of each department according to the causal relationship of business logic into an evidence-based chain of pain points, thereby forming a clear business process diagram. For example, the business process of an enterprise producing apple juice can be logically clustered as follows: first, the "planting procurement transportation" of upstream raw materials; Secondly, the intermediate production process involves "scheduling filling packaging quality inspection"; Thirdly, the "shipment logistics marketing channel management" of downstream sales is further explored and refined layer by layer based on this logical clustering, forming a multi-level perspective of the business process.
On the business flow perspective, the data flow formed by the overlapping enterprise information system data collection corresponding to each node is the basic prerequisite for digital transformation. This data flow with business logic architecture perfectly maps from the static structure of enterprise organizational structure/business units to the dynamic collaboration of business operations, helping enterprises to follow the map, locate and quantify the causes and effects of various pain points in the evidence-based chain, and thus determine which pain points have the feasibility of backup digital technology solutions.
Secondly, business experts should actively lead the manifestation of pain points, rather than passively responding.
Many business experts in enterprises tend to have a preconceived distrust of experts from other fields, believing that the dataization of business flows will distract energy and slow down the current main business. In fact, business experts can become proactive leaders in digital transformation, leveraging their familiarity with business process logic and combining it with the early IT construction of the enterprise to promote the integration of various IT systems' content (i.e. data) into a data flow according to business flow, thereby achieving business dataization and driving the digital transformation to take off. Because based on massive data streams, various data technologies such as Data may be appliedTechnology（DT）探索解决痛点的规律。
技术锚点价值化——新价值而非新技术为导向
当前人工智能（AI）、deep learning、big data和云计算等新技术名词很流行，企业选择数字化转型的技术路径也容易陷入到“追星”的陷阱。
其实，最新技术或者学术明星，并不一定是企业数字化转型的最佳选择，原因有二：其一，最新技术往往在特定的先决条件下才可能发挥作用，在具体业务中的适用性与稳定性有待验证，如同每一种新药上市前都要广泛的临床试验；其二，具体业务场景中的挑战与学术研究大为不同，学术明星擅长在前人研究基础上找到创新点，但不保证创新点一定能够带来足够大的实战价值，而具体业务场景里应用新技术则需要有最低价值门槛，至少投入产出比要足够。
而如何甄选合适的新技术作为锚点，可以从以下两个维度分析：
第一，采用ROI（投资回报率）相对高的技术。
对于选定数字化转型拟解决的痛点，企业的业务专家可以给出期待的收益价值，技术服务商提出的解决方案实施成本则包括开发实施成本和企业内部业务线的配合成本，基于收益与成本的考量、按照投资收益比ROI，综合选择适合的技术方案。实践表明新鲜出炉的明星理论往往有相当长的开发试错周期，导致转型项目的夭折。
第二，采用嵌入式成本相对较低的技术实施方式。
大部分的数字化转型项目都需要把新技术与现有的系统做某种程度上的兼容对接，所以应当着重考虑新技术的嵌入式成本，在保证新技术向后兼容的同时，应避免技术选型与现有系统之间产生无法共存的排斥反应，尽最大可能实现“无缝对接”。
例如，某新技术要求数据源A实现每秒更新一次，而数据源A的采集系统一直是每分钟更新一次，强行上马则可能造成采集系统的崩溃，显然这是一个嵌入式风险高的选项。2018年阿里云工业大脑团队实施恒逸石化项目，当时目标是通过AI提高己内酰胺锅炉燃烧效能。如果想要追求极致，工厂希望实现自动反向控制，即工人不需要操作，生产线能够自动根据算法的结果调节工序关键参数。但这种方式需要打通现有系统，集成成本过高，还可能有未知的控制风险。所以最终采用了折中的解决方案：将AI计算的参数，即时推送至业务流程中，再由业务操作者来决策是否应用该参数。这种方案减少了嵌入式成本并且降低了风险，最终实现了方案快速上线，提升了燃烧效率2.6%。
转型拐点连续化——从量变到质变的渐进过程
数字化转型一定是持久战，而非毕其功于一役。实践中，一些企业满怀热情启动数字化转型项目，在成功完成一两个项目后就开始做财务核算，判断是否要追加投入继续做大。而此时最容易陷入用成人百米赛跑成绩选拔少年选手的误区，只看到眼前而忽略了蕴藏的潜力。
评估前期的数字化转型项目成功与否，不仅要看其创造价值是否明显，更要关注其方式和路径是否能再度放大持续创造新价值。
例如，在2018年阿里云与海底捞及蒙牛的数字化转型案例中，首期项目以数据中台的雏形为基础，然后开发了一两个示范应用。如果通过示范应用的业务价值了解数据中台的威力，继而加码投入，则会创造出更多的示范应用，解决业务痛点甚至是创造新的业务模式，用数字化技术打造连串的价值创造点，从而触发质变的转型拐点。
阿里云城市大脑在交通领域的实践也经历了类似的过程，从2017年在杭州中河高架及萧山区试点路口到2018年拓展至杭州主城区1300个路口420平方公里区域，站在2017年与2018年底分别评价城市大脑的收益，得到的结论显然有了巨大的飞跃。
传统产业的数字化转型之路漫漫，成功的路径不可能复制，因为各家企业的现状迥异，但是选择成功路径的方法论是可以借鉴的，“三点一线”就是多个企业实践的提炼的方法论，希望对还在求索中的前行者有所启迪。
信息来源：电子发烧友网
Editor He Yang of this website/Organizing
