When it comes to data, everyone likes to talk about big data, precision marketing, customer management, and corporate insights, but in fact, for most small and medium-sized enterprises, the data accumulated from various channels over the years since its establishment is integrated. , and only small data.
These data, with the right tools, can be easily managed. However, before choosing a tool, you can ask yourself this question: What kind of analysis will I use with this enterprise data?
FanRuan data report
Using the logic of data analysis, two dimensions are also set for "how to analyze enterprise data": value and complexity. According to different degrees, it can be divided into four types: descriptive, diagnostic, predictive and instructive.
[Descriptive] analysis answers the question of what: What happened to my business?
Data types are often comprehensive, extensive, real-time, accurate, and presented in efficient visualizations.
For example, a company's monthly sales report is a [descriptive] analysis.
A Kanban board summarizes the monthly indicators, completion rates, and completion degrees of various regions. It changes in real time and is automatically aggregated by the end of the month. It is not only "description", but also a certain degree of analysis, which can meet the daily management needs. For example, in Yangzhou, the target completion rate for this month is the lowest, but the annual target completion rate is relatively good. Is it because the sales target for this month is too high, or the assessment is lax. If it is artificial slack, should the annual assessment also record the assessment results of reading?
Efficient visualization, on the one hand, it means that the speed of making this report is faster, and you can answer questions immediately. No, I want to know how the situation is today, but you will tell me the answer after three days; on the other hand, this report is based on The form of "template" exists. When the data changes, the report will also change accordingly. When it is opened, it is always up to date.
The [diagnostic] analysis answers the why question: Why is this happening to my business?
-
Requires the ability to drill down from the big picture to the details
-
Requires the ability to isolate all obfuscated information
For example, looking at the data map, I found that the market sales in Jiangsu are relatively high. I want to know why? So click on the province, you can locate the sales data of various products and the corresponding cooperative customer data.
FanRuan Data Decision System
[Predictive] analytics answer what likely questions: What is going to happen to my business?
主要回答战略性的问题：我的商业策略是否在一段时期内保持一致，根据算法，用历史模于预测某个具体的结果。
就像玩儿三国杀的时候，很多人喜欢诸葛亮，不停地使用“观星”一样，我们希望能够预测某件事在未来发生的可能性，或是预测一个可以量化的值，甚至预测某个结果可能发生的时间点。如何达成预测，一方面取决于你的工具，但更重要的，取决于你的预测模型。
所有的工具用到最后，拼的都不再是工具，而是用工具的人。
就像同样一台相机，马格南和司马南，用出来效果肯定不一样。
［指导型］的分析，回答的也是一个what的问题，但这个问题很直接：我需要做些什么？
最后，基于你已经知道的“发生了什么”、“为什么会发生”以及“未来可能发生什么”的分析，［指导型］分析可以帮助你确定可以采取的措施，也就是：驱动行动。
虽然我们习惯地称后两种分析为“高级分析”，但是比较客观地说法是：不同类型的分析能提供不同的商业价值，每一种分析都有它自己的用处。对于中小企业团队来说，80%对于数据的需求，都集中在［描述型］分析与［诊断型］分析之中。那么，要如何管理数据，才能更高效地利用企业数据完成［描述］与［诊断］呢？
管理数据的方法：
1、弱水三千，先取三千，再取一瓢
中小企业的数据缺乏系统化、规模化的管理，企业数据通常散落在各个部门手中，然后每个部门的数据，散落在不同业务人的手中。若想对企业的全景做出［描述］甚至［诊断］性分析，所要做的第一步是，把所有的数据集中管理——此为“弱水三千，先取三千”。
集中在一起之后，砍掉冗余数据，统一数据格式与粒度，梳理有业务需要的指标体系，形成CEO管理看板、财务管理看板、销售管理看板、市场部管理看板、人力资源管理看板等等。每个人看到的，都是跟自己工作密切、直接、实时相关的指标，这叫做“再取一瓢”。
FineReport数据决策系统（样本）
2、突破IT瓶颈，学会自问自答
Managers have their own management kanban, which may look at a relatively macroscopic aspect. If a problem is found, drill down and analyze it to make the work more efficient. However, if business personnel want to improve their business, it is impossible to be satisfied with a fixed Kanban. The best way is to work together with [accurate data authority] + [self-service analysis tools], so that business personnel can break through IT bottlenecks and use the data in their hands to ask and answer their own questions.
[Precise data authority] The breakthrough is the ownership of data; I am in charge of my business data, except for tampering, I should be able to see them at any time and use them for analysis, rather than waiting for the "handout" of the IT department.
[Self-service analysis tool], the breakthrough is the degree of use of data, I can analyze my business data myself, and I can analyze it according to my own perspective, and even revise my ideas while analyzing, instead of writing requirements. IT personnel help analyze, and if you want to change the demand, you have to ask for it, and wait patiently.
There are many self-service BI tools. For example, when I use FineBI to analyze, my IT colleagues help me organize the data into the business package. When I need to analyze the data, I go to the business package to find it. The data will be updated every night at 12:00. to ensure the timeliness of the data.
3. Unified Cognitive Chemistry as Cooperation
Finally, I think that what small and medium-sized enterprises are fighting for is vitality, immediate response, and full cooperation. The company is already very small, so there is no need to force the sales department and the marketing department, the front-end and the back-end, human resources and administration. Speaking with objective data, unifying the cognition of the current situation of the enterprise and the reasons behind it, it will be better to spend the time used to tear up, throw the pot, and throw the monkey on the solution and action.