Business intelligence BI has gained multiple recognitions in terms of reputation and share in the market, so what exactly is business intelligence BI? What exactly does business intelligence BI mean? Why is business intelligence BI so popular? Today, we will address these issues and take you through a detailed understanding of what business intelligence BI is and what its implementation means to enterprises.
1. What is business intelligence BI?
Business Intelligence BI - Parker Data Business Intelligence BI Visual Analysis Platform
The implementation of various informatization and digital technologies and applications has given people a deeper understanding that data is the "oil" of the future. Data is similar to oil. It needs to be developed before it can truly reveal its huge value. This is why current enterprises apply data analysis and data processing, and it is also why business intelligence BI has such a hot demand.
Some people must be curious after seeing this, why does business intelligence BI have a relationship with data, and what is its significance after it is implemented in the enterprise? To briefly summarize, business intelligence is a complete set of data technology solutions consisting of data warehouse, query reports, data analysis, etc., which can realize the standardization, processization and standardization of business processes and business data, and connect ERP, OA, Different business information systems such as CRM integrate and summarize enterprise data.
Some people may have heard that although business intelligence BI has high value to different groups of people in the enterprise, those who need business intelligence BI the most are the senior managers of the enterprise. They use business intelligence BI and data visualization to meet the needs of different groups of people in the enterprise for data query, analysis and exploration, and to obtain information and knowledge, thereby providing data basis and decision-making support for management and business.
To summarize the core content of business intelligence BI, there are roughly three major characteristics:
- A complete set of data technology solutions consisting of data warehouse, query reports, data analysis, etc.;
- Connect and effectively integrate data in different systems (ERP, OA) in the enterprise;
- Use appropriate query and analysis tools to quickly and accurately provide reports to provide decision support for enterprises.
Business Intelligence BI - Parker Data Business Intelligence BI Visual Analysis Platform
Back to the topic, we have learned what business intelligence BI is and what functions enterprises can achieve through business intelligence BI. So, does the implementation of business intelligence BI in enterprises have any deep meaning? When talking about the meaning of business intelligence BI, What should be understood.
In fact, business intelligence BI is indeed very important in the enterprise. This is because business intelligence BI plays a connecting role in the enterprise. Looking down, business intelligence BI can open up different business information systems such as ERP, OA, and CRM, and will clean The final data is stored uniformly in the data warehouse; looking up, business intelligence BI can provide data visualization reports of different themes and forms, comprehensively display the development status of the enterprise through data visualization analysis, and assist managers in making decisions.
Business intelligence BI can be divided into three levels according to different stages of the enterprise data life cycle:
The first layer, the visual analysis presentation layer - the demand layer of business intelligence BI, on the one hand represents the user's needs, what the user wants to see, what they want to see, on the other hand it also represents what the user wants to analyze, these are on this layer Show.
The second layer, the data model layer - business intelligence BI data warehouse, is mainly responsible for the analysis model of enterprise data and completes the transformation from business calculation rules to data calculation rules.
The third layer, data source layer - the data layer of business intelligence BI, business information systems of different departments and business lines, the data of the underlying database is extracted into the data warehouse of business intelligence BI through ETL, modeling analysis, etc., and ultimately supports Visual analysis display on the front end.
2. The position of business intelligence BI in enterprise IT informatization
As mentioned earlier, the role of business intelligence BI in an enterprise is mainly to connect the past and the next, and is an important part of the information construction. Business intelligence BI forms a complete set of data value system around data, giving full play to the role of data in the enterprise.
When talking about the position of business intelligence BI in enterprise IT informatization, we must first understand what the enterprise's informatization construction is. Generally speaking, the informatization construction of enterprises is universal, so the IT informatization of most enterprises can be divided into two stages: one is business informatization, and the other is data informatization.
Enterprise informatization-Picodata business intelligence BI visual analysis platform
Business informatization - ERP, CRM, OA and self-built business information systems used by enterprises are collectively referred to as business informatization. The main function of business informatization is to optimize and adjust an enterprise's business processes. Through standardization, standardization, and onlineization, it can improve business operation efficiency, reduce enterprise manpower, time, energy and other costs, and accumulate a large amount of business data, etc., which is the basis of business management ideas. It reflects the modern way of business management.
Data informatization - Big data, business intelligence (BI), data analysis, data mining, etc. that we often hear are all collectively referred to as data informatization. Data informatization can help enterprises comprehensively understand their operations and management, adjust their operation and management models from experience-driven to data-driven, reduce subjective effects such as emotions and psychology, form data-based business decision-making support, and improve the accuracy of decision-making. This is a higher-level corporate management method for enterprises.
Enterprise informatization-Picodata business intelligence BI visual analysis platform
The information construction of an enterprise is a complete process. Without the construction of business systems, there will be no data accumulation. Without data precipitation, enterprises will have no basis for deploying business intelligence BI. This is the two-way effect of business informatization and data informatization, which allows business systems to promote the deployment of business intelligence BI, and also allows business intelligence BI to improve the effectiveness of business systems.
3. Who are the main users of business intelligence BI?
Business Intelligence BI - Parker Data Business Intelligence BI Visual Analysis Platform
The main users of business informatization - The main users of business informatization are front-line business personnel, so the users of business informatization are more from a business perspective, entering data, recording processes, viewing business information, etc. for the business.
The main users of data informatization - The main users of data informatization are management decision-makers. In daily processes such as business management, decision-makers use data technology solutions such as business intelligence (BI) to position themselves more from a management perspective. problems, analyze problems, and ultimately form business decisions.
4. What exactly does the data island mean?
The term data island is actually very vivid. From the literal meaning, we can understand that data island means that data is isolated from each other like islands and cannot communicate and collaborate with each other. As a result, it is difficult for enterprise managers to obtain comprehensive data about the enterprise when they understand enterprise information and make decisions, which in turn makes it difficult for middle and senior managers to obtain data from different departments. This not only leads to problems in decision-making , it may even make the enterprise unable to handle abnormal data, unable to solve the discovered problems, and put the enterprise into trouble.
Business Intelligence BI - Parker Data Business Intelligence BI Visual Analysis Platform
As a data technology solution, business intelligence BI can solve the "data island" and "information island" problems faced by enterprises through data informatization, data warehouse and data visualization when faced with the problem of data islands. Therefore, business intelligence BI requires enterprises to Senior managers carry out planning and mainly provide decision-making information to managers at all levels of the enterprise to assist in decision-making.
Management Cockpit - Parker Data Business Intelligence BI Visual Analysis Platform
Therefore, when introducing business intelligence BI, it is necessary to understand the needs of different people. From the perspective of different employees in the enterprise, some people believe that data islands exist and must be resolved. Some people do not think that data islands exist. Even if they exist, it will have no impact on them, so there is no need to solve them. The fundamental reason is that they do not grasp the real service objects of business intelligence BI.
5. How business intelligence BI obtains data from the business system
Business intelligence BI obtains data by accessing and connecting the business system data source database. No matter what type of database it is, business intelligence BI connects the database through ETL to extract the original table data of the business system and processes it in the data warehouse. Finally, Supports front-end visual analysis report display.
Business Intelligence BI - Parker Data Business Intelligence BI Visual Analysis Platform
A friend asked this question before: Does the data source layer need to develop an interface?
In fact, generally speaking, it is not necessary. Basically, those who ask this question have experienced the interface docking of software systems. The interface docking of software systems is because some business software is developed by JAVA, some by .NET, and some by B/S. Architecture, some are C/S architecture.
The interface between software systems requires development participation, mainly to connect the business processes of different software. This kind of interface requires moving code. However, the interface for obtaining data in business intelligence BI is different. It has nothing to do with the business system software itself. It only needs to access and connect to the database behind the business system, and directly obtain data from the database, so there is no need for a software interface, or There is no software interface to access this concept.
Except in one case, the business system is a public cloud and pure SAAS mode. In this case, the data can only be accessed through the API interface open to the outside world by the software.
Performance analysis of sales personnel in a certain pharmaceutical industry - Paco Data Business Intelligence BI Visual Analysis Platform
Channel terminal management analysis of a certain liquor industry-Pakdata Business Intelligence BI Visual Analysis Platform
6. How should we understand the relationship between data center, business intelligence BI, and big data?
When the system's business intelligence BI encounters large data volumes and unstructured data processing scenarios, the underlying data warehouse is upgraded to a big data data warehouse architecture. This is business intelligence BI analysis under big data; in the data of big data On the basis of the warehouse architecture, the data collection capabilities are further expanded to the left. In the middle, in addition to the data warehouse modeling of the original big data architecture, the concept of data assets, data asset inventory, and data asset management are added. To the right The data service capabilities have been expanded, and the data processed in the data center according to certain rules are packaged and provided to external services. Therefore, data collection, data warehouse modeling, data asset management and data services under the big data architecture constitute the core of the data center.
Data Visualization - Parker Data Business Intelligence BI Visual Analysis Platform
The foundation of the data middle platform is the big data architecture. The data warehouse is a big data upgrade of the traditional business intelligence BI data warehouse, and the business intelligence BI becomes the application layer above the data middle platform, using the data services of the middle platform to obtain data for processing. Analysis shows.
This is the relationship between business intelligence BI, big data, and data center and the evolution process in different data scenarios and service scenarios. If you understand this process clearly, you should not easily confuse their concepts. As for which business intelligence BI, big data, or data middle platform should be chosen? In fact, how to choose the appropriate technical route and technical architecture ultimately depends on what the company itself wants to solve, and cannot be chosen blindly. The result of blind choice is large investment and small output that does not meet the expected expectations. We should still focus on the demand itself, demand is king.
7. Several misunderstandings about business intelligence BI cognition
Many companies use business intelligence BI as a pure reporting tool, and the output form becomes a visual chart. However, the content displayed in the chart is still the previous department business information. It only shows the basic situation of the front-line business department, and managers still need to spend a lot of money. Spend time and energy to understand the overall development of the company.
Business Intelligence BI - Parker Data Business Intelligence BI Visual Analysis Platform
I have summarized here some misunderstandings that people often encounter in their understanding of business intelligence BI:
1. Business intelligence BI is report visualization, which is a bunch of visual charts. Business intelligence BI is front-end visualization.
2. Business intelligence BI is a drag-and-drop analysis tool product.
3. Business intelligence BI is business intelligence BI and has nothing to do with data warehouse.
4. With business intelligence BI, there is no need for data warehouse modeling. Business personnel can do business intelligence BI analysis by themselves, and they can drag and drop to do business intelligence BI analysis.
5. Business intelligence BI is business-driven and does not require support from IT personnel. Agile business intelligence BI does not require IT intervention.
6. Isn’t direct connection to business intelligence BI good? Analysis can be done by directly connecting to the data source and does not require a data warehouse.
First, let’s briefly correct our understanding of these issues.
1. Business intelligence BI is report visualization, which is a bunch of visual charts, and BI is front-end visualization.
Business intelligence BI is a complete set of data technology solutions consisting of data warehouse, data analysis, data reports, etc. In a BI project, 20% of the time is spent on front-end analysis reports, and 80% of the time is spent on the underlying data Warehouse design, ETL development, data retrieval development, etc.
Therefore, the visual report is only the final presentation of business intelligence BI, but it is not the whole of business intelligence BI.
2. Business intelligence BI is a drag-and-drop analysis tool product.
To be precise, drag-and-drop visual analysis tools can only solve part of business intelligence BI, that is, visual analysis. But in fact, the technical scope of business intelligence BI is relatively wide, involving all aspects from underlying data acquisition to front-end display and analysis.
Strictly speaking, simple drag-and-drop business intelligence BI visual analysis tools can only be positioned at the individual and department levels, which are very different from enterprise-level business intelligence BI. Therefore, a simple business intelligence BI analysis tool cannot play a role in business intelligence BI. The real role of BI cannot replace the position of business intelligence BI.
3. In the past, some people always said that business intelligence BI is business-driven, business intelligence BI is BI, and has nothing to do with data warehouse.
This question is very profound. I have thought so in the past. I always felt that with business intelligence BI, there is no need for data warehouse modeling. Business personnel can do business intelligence BI analysis by themselves, and they can drag and drop to do business intelligence BI analysis. It does not require support from IT personnel. Agile business intelligence BI does not require IT intervention and does not require the construction of a data warehouse.
Management Cockpit Large Screen - Paco Data Business Intelligence BI Visual Analysis Platform
Any business intelligence BI sales or pre-sales tell users that your company's business intelligence BI project does not need to build a data warehouse. All analyzes in the company can be done directly through drag and drop of the business intelligence BI analysis tools, without the need for IT staff support. , business personnel can do it by themselves... Anyone who dares to make such a promise either does not understand business intelligence BI, or is really cheating.
In the construction of enterprise-level business intelligence BI projects, it can really be achieved by business personnel simply dragging and dropping to achieve data visualization analysis. At least in my personal work experience of more than ten years, more than 95% of companies have Can't do it. Key companies I have served include: SHP (Security Health Plan), Microsoft (China), Microsoft (USA), VWFC (VW Financial), etc.
VWFC does a very good job. A few business personnel make many reports by themselves and run thousands of reports online. Why? Because the underlying data warehouse has been built for many years, the underlying data architecture is relatively standardized. Business Driven Business drive, what is its premise?
1) The quality of the underlying data is very standardized, the data warehouse structure is very complete, and business personnel are not allowed to touch the underlying data. ETL, data retrieval, indicator calculation, etc. are all maintained by the IT department.
2) Business personnel must be proficient in the use of business intelligence BI front-end reporting tools through training, and must understand the released data analysis model interfaces.
3) Business personnel must be very familiar with business and data.
Many companies do not have problems with items 2) and 3). Can item 1) be solved by directly building a front-end business intelligence BI tool for business personnel to solve? Obviously business personnel do not have this ability.
This is why when it comes to training, the use of business intelligence BI tools is very simple, but once it comes to actual enterprise business intelligence BI project development, it is found to be difficult to implement. Because during training, the data tables given are all selected and are always of high quality and standardized. They only require simple left tables and right tables, such as sales order tables and order details tables. Naturally, it is easy to give visual reports to Realize it.
Data Visualization - Parker Data Business Intelligence BI Visual Analysis Platform
However, in actual enterprise business intelligence BI project analysis, the calculation rules of analysis indicators cannot be solved by simply associating a few tables. If you don’t believe it, you can challenge an actual indicator calculation logic: Challenge a small case of ETL data cleaning in the database It is very simple to understand the data with just a data table, but many business intelligence BI developers also need to waste a lot of energy to do it, let alone self-service business intelligence BI analysis by business personnel.
I am not talking about this to blindly deny the role and capabilities of self-service business intelligence BI. Self-service business intelligence BI has its usage scenarios, and it has indeed helped us simplify a lot of BI work. However, from a professional perspective, I am particularly disgusted with this part. Business intelligence BI vendors have repeatedly reinforced concepts similar to this to the market in an irresponsible way: business intelligence BI is visual reporting, business intelligence BI does not require data warehouse modeling, traditional data warehouse modeling is very backward, business intelligence BI It is self-service analysis and business intelligence BI. Self-service analysis is very simple. Business users can learn it in just a few days of training and analyze how they want...
From the perspective of marketing and sales, there is no problem with publicity that simplifies the complexity and difficulty of getting started with the product. What is problematic is using a wrong explanation and unprofessional explanation that ultimately misleads companies into accepting these incorrect concepts. , and use these incorrect concepts to evaluate and plan the construction of business intelligence BI projects, without fully anticipating the challenges and risks that may be encountered during the construction of business intelligence BI projects, which ultimately leads to unsuccessful projects and repeated construction. .
We have a customer in Beijing who spent more than one million yuan on a so-called business intelligence BI project. The project went online for about a year, but in the end it could not be pushed forward at all and failed. Later, we found Pico Data, and we introduced them to the Pico Data business intelligence BI analysis platform. We did this project for several consecutive periods, and the customer even wrote a letter of thanks. Why it couldn't be pushed forward and the project failed before: not paying attention to the planning of data warehouse. Because their business is continuous and changing, annual demands need to be dynamically adjusted, data continues to increase, and the depth and breadth of analysis are constantly changing. Without a good underlying data architecture to support it, they rely solely on SQL to obtain data. It is impossible to build data sets and generate reports in the form of data to support the changes in business analysis needs of an enterprise in the next 3-5 years or even longer.
In addition to this case, there are many records on my mobile phone of business intelligence BI that I have tried before but failed and did not do well. I came here to chat and complain. Is the product really bad? I also objectively helped them analyze: some of these products are in Gartner's Magic Phenomenon Leader quadrant. Do you think the products are good? Some products are old brands that have been in the domestic business intelligence BI field for many years. Do you think the products are good? Objectively speaking, from my personal point of view, these products are actually excellent, and there are no big problems with the products themselves.
The problem is that so many companies that need business intelligence BI from scratch don’t know that there are so many pitfalls in a business intelligence BI project. Will many business intelligence BI vendors explain these points clearly to enterprise customers? How to do a business intelligence BI project, what kind of risks are involved, what kind of problems you will encounter in the future, how to solve these problems, what kind of methodologies and means... If it is just to sell a set of business intelligence BI Products or tools, do you think these business intelligence BI salespeople will tell their customers about these things? No, at least it won't be too in-depth or comprehensive, because it makes the difficulty of business intelligence BI too complicated. Once it is not explained well, it will reduce the trust of customers.
Sometimes we don’t talk about it because we are afraid that it will be complicated and the decision-making cycle of corporate customers will be too long. Sometimes we don’t talk about it because we don’t understand. If you don’t tell them, customers won’t know and have no experience, and problems will arise in subsequent business intelligence BI project construction.
At a conference, a senior pre-sales technical expert from a business intelligence BI vendor said something when communicating with customers: Isn’t direct connection to business intelligence BI great? Analysis can be done by directly connecting to the data source and does not require a data warehouse. Ignorant people are fearless. I really couldn't listen anymore, so I interrupted and communicated directly. Through communication, it can be judged that this so-called technical expert basically has no experience in completing a complete business intelligence BI project, and his ability to build a business intelligence BI project from scratch is zero. It is very difficult to guide a business intelligence BI project with customers with such ability. Can the quality of this business intelligence BI project be guaranteed?
This is the reason why we, Paike Data, and I personally created the video account "Lv Pin Chat Data" to objectively talk about business intelligence BI, objectively talk about data, and popularize the knowledge and concepts of business intelligence BI that we think are correct. Tell us the vast number of business intelligence BI users, how should they understand and recognize business intelligence BI, and what pitfalls does business intelligence BI have that our companies need to pay attention to.
We cannot say that the knowledge and concepts we talk about in the field of business intelligence BI at Parker Data are necessarily universally applicable, but we welcome any business intelligence BI vendors or any individual BI enthusiasts to share some knowledge and ideas about business intelligence BI. The concept challenges us to see if some of the business intelligence BI knowledge concepts popularized by Parker Data are correct. If it is popularized correctly, it means that everyone has indeed encountered these problems, and these knowledge and concepts are rare experiences for enterprises. If the popularization is wrong and what is wrong, point it out and let’s take a look and discuss what else we can do for the enterprise.
8. Where did the reporting tool come from?
I have been in the technology field, information field, and business intelligence BI industry for more than ten years, and I have never left this circle. Developed technologies such as JAVA (AWT, SWING, JSP, Hibernate, Spring, ibatis), .NET (ASP, http://ASP.NET , C#.NET), Object-C, JS, etc., and developed business software system platforms .
In the early days, front-end technology was very weak, and the implementation of AJAX also required handwriting. To achieve click-to-edit and modify data in a form, you needed to use JS DOM yourself. Making reports is basically JSP or ASP scripting language nested in HTML on the front end for loop output. The report style is very primitive and ugly. Slightly more complex tabular report styles need to be adjusted with JS.
The reports used at that time, such as Crystal Report, Runqian Report, etc., had tags in the front-end scripting language that could be directly referenced, and report generation replaced a large amount of handwritten code. In the early days, front-end and back-end technologies were not separated. ASP.NET was slightly better. The front-end gradually had some integrated controls that could be used directly, but JAVA did not. When did this stage mentioned above happen? Around 2005 or 2007, I think it was already widely used. You should still be able to find many original report tag posts on the old CSDN.
Older reports such as Jinfeng Report Jreport, Star Report StyleReport, etc. also have a certain market in China. As early as 2010, the revenue of some report manufacturers had exceeded 100 million, which shows that the market for basic reports is still very good.
What was the report positioning at that time? It was a pure Report report. The data aggregation list returned from the back-end database query through the program was then bound to the front-end script page to generate various reports. In fact, it was used in various business software. The report display in the system is far from the level of business intelligence BI analysis.
And there are a large number of software developers that actually have strong reporting capabilities, but these reporting capabilities are not separately available as reporting products on the market.
Gradually, with the improvement of front-end technology and front-end framework, traditional table technology has moved on to the visual display of various column charts, bar charts, and pie charts. At this stage, the boundaries between reports and business intelligence BI are becoming increasingly blurred. . Why? The report presentation capability of business intelligence BI is roughly equivalent to that of traditional reports. There has not yet been the ability to achieve rapid multi-dimensional analysis through self-service analysis and self-service drag and drop.
The main thing I want to say after talking so much is that many business intelligence BI projects we have seen are implemented with report thinking, that is, SQL is used to display data sets on the front end. And what should the real business intelligence BI thinking be? Multidimensional thinking and model thinking determine the final direction of a business intelligence BI project. These points will be discussed in detail later.
9. The essence of business intelligence BI - the implementation of enterprise business management thinking
The Essence of Business Intelligence BI - Parker Data Business Intelligence BI Visual Analysis Platform
What exactly is business intelligence BI? technology? product? Or something else? We have raised our understanding of BI to another level: business intelligence BI is the implementation of an enterprise's business and management thinking. How to understand this? To put it simply, the content presented on the visual report is what a company is really concerned about. There are analytical indicators of the company's operating performance that top management focuses on, as well as those of a specific department.
10. What are the differences and connections between business intelligence BI and data warehouse Data Warehouse?
We often encounter people asking what is the difference between business intelligence BI and data warehouse. In fact, this question reflects that some friends’ understanding of business intelligence BI is still somewhat inaccurate and biased. This question actually conceptually separates BI and data warehouse. The data warehouse is artificially fragmented. This situation is actually quite normal, because everyone’s first impression of business intelligence BI is all kinds of cool visual charts and reports. In addition, there are many lightweight front-end visual business intelligence BI analysis tools on the market. The understanding of BI only stops at the visualization part.
To be precise, business intelligence BI not only includes front-end visual analysis and report presentation capabilities, but also includes the construction process of the underlying data warehouse. Gartner already mentioned Business Intelligence in the 1990s. It believes more that: BI is a data technology solution that extracts analytically valuable data from many different enterprise business systems and cleans it. , transformation and loading, which is the ETL process of Extraction, Transformation, and Loading, and finally merged into a data warehouse according to a certain modeling method, such as Inmon's 3NF modeling, Kimball's dimensional modeling, or both. Based on the hybrid architecture model, appropriate analysis and presentation tools are finally used to form various visual analysis reports to provide data decision-making support for the enterprise's management decision-making layer.
Business Intelligence BI - Parker Data Business Intelligence BI Visual Analysis Platform
Therefore, you can see from here that the location of the data warehouse is the layer between visual reports and underlying business system data sources. It plays a connecting role in the entire business intelligence BI project solution. If business intelligence BI is compared to a person, the upper body, especially the face, is the appearance, the lower body is down-to-earth and absorbs the essence of the earth, and the core and core strength of the middle part are the data warehouse.
Then everyone will also ask, aren't there many business intelligence BI tools on the market that can directly link to data sources and perform drag-and-drop analysis? Can't they also make business intelligence BI analysis reports? This kind of independent, front-end oriented business intelligence BI analysis tool is more positioned as a department-level and individual-level business intelligence BI analysis tool. It is suitable for many scenarios that require deep-level complex data processing, integration, modeling, etc. It cannot be solved. The best way is to build a complete data warehouse at the bottom, standardize many analysis models, and then combine these front-end business intelligence BI analysis tools. Only in this way can the front-end business intelligence BI analysis capabilities be truly released.
Many companies think that just buying a front-end business intelligence BI analysis tool can solve all enterprise-level business intelligence BI problems. This view is actually not feasible. This type of business intelligence BI analysis tool may not be a problem when the initial analysis scenario is relatively simple and the complexity of connecting data is not very high. However, the construction of business intelligence BI projects in enterprises has a characteristic, which is a spiral construction process. Because there may be more and more business systems being connected, the depth and breadth of analysis will be increasing, and the complexity of data will become more and more challenging. At this time, there is no good data warehouse architecture support, and relying solely on Front-end BI analysis tools are basically impossible to handle.
Data Warehouse - Parker Data Business Intelligence BI Visual Analysis Platform
Just like going to a traditional Chinese medicine store to get medicine, the reason why it is so fast is because before getting the medicine, others have already sorted out the various native Chinese medicinal materials (data from the original data source), cleaned them up and put them away, so that they can think about how to match the medicines (dimension Visualization of indicator combinations) is very fast.
There are many such companies in China, and the lack of depth of understanding of business intelligence BI has led to some directional errors in the construction of business intelligence BI projects. In the end, it has been difficult to continue to advance the business intelligence BI project.
Therefore, in enterprises, we need to clarify whether our business intelligence BI construction is for enterprise-level or individual and departmental analysis work. If you are a personal data analyst, it is enough to use such front-end business intelligence BI analysis tools. If you need to build an enterprise-level business intelligence BI project, you should not only focus on the front-end visual analysis capabilities, but also the construction of the underlying data architecture, that is, the data warehouse level.
11. Data warehouse modeling methodology Kimball vs Inmon and hybrid architecture
Data warehouse modeling is the top priority in the construction of business intelligence BI projects. Inmon's three paradigm 3NF modeling and Kimball's dimensional modeling are both business intelligence BI data warehouse modeling methodologies. These two business intelligence BI modeling What are the differences and connections in the ways.
12. Methodology for implementing requirements when actually carrying out a BI project
Business intelligence BI is completely demand-driven. Since it is a demand, interviews and research are required. Before conducting interviews and research on business intelligence BI needs, you need to be familiar with the business characteristics of the industry in advance. Based on the company itself, you need to be familiar with their business processes, as well as the key points that they will probably focus on in the departments you are interviewing. All these need to be sorted out in advance. Establish the entire business framework in your mind and practice it repeatedly.
13. What kind of enterprises should adopt business intelligence BI?
What kind of enterprises are suitable for business intelligence BI? Look at the degree of basic business informatization and the level of detail and granularity of daily business management. The degree of basic business informatization refers to the basic construction of the enterprise's own IT business system. Without the support of the business system, there will be a lack of data foundation for business intelligence BI. The second is the granularity of business management. Is the level of the enterprise's own business management more detailed? There is an urgent need to improve the efficiency of business management and decision support through business intelligence BI.
14. How to efficiently provide BI data analysis reports and summaries to senior leaders
After completing the business intelligence BI project, you must also consider how to report to the boss in the end, and master the business intelligence BI data analysis thinking framework and the five key points of reporting: user business level and scope, work results, plan execution review, problem feedback, Outlook planning and vision.
Business Intelligence BI - Parker Data Business Intelligence BI Visual Analysis Platform
This is just a simple reporting framework, and there are many points that can be added to it. For example, focus on the industry and talk about how industry driving factors are combined with business intelligence BI; from the perspective of corporate operation and management, how corporate vision, CSF, KPI and performance are decomposed and reorganized; such as attribution analysis from a financial perspective; pyramid management Models; dynamic indicator library composition principles, etc. can be selectively integrated and explained.
15. The integration of business intelligence BI and enterprise management
Business intelligence BI analysis is highly integrated with enterprise management analysis. Enterprises with high ROE may have high profits, such as Moutai and the jewelry industry, may have fast turnover, such as the retail industry, or may have relatively strong financing capabilities and use leverage. ROE attribution analysis looks at industry characteristics.
16. Accumulation of industry and business knowledge for business intelligence BI projects
To do business intelligence BI, you must also be familiar with industry and business knowledge. Without combining industry business knowledge, it is difficult to implement business intelligence BI projects. The essence of business intelligence BI is actually the implementation of an enterprise's business and management thinking. Why do corporate executives and business department managers use business intelligence BI to read reports? What do they look at and what do they focus on? These contents are the focus of their daily business operations and management in the enterprise.
Data Visualization - Parker Data Business Intelligence BI Visual Analysis Platform
The seemingly scattered reports in business intelligence BI projects actually have strong logical correlations in the eyes of actual users. And the higher-level managers look at the content of business intelligence BI reports, the more focused they are on business results. People in front-line business departments may pay more scattered attention and look at detailed business process data.
Therefore, an excellent business intelligence BI developer or development consultant not only needs to be polished at the technical level, but also needs to accumulate industry knowledge and corporate business knowledge.
17. Topics on real-time processing of business intelligence BI
Business intelligence BI has a certain lag in data processing, and usually adopts T+1 mode. The main reason is that the ETL data processing process requires a lot of time loss, and usually uses space for time.
Designing the ETL schedule that was previously layered according to the business intelligence BI data warehouse into a schedule that can automatically find dependencies based on individual indicators greatly increases the flexibility of individual indicator scheduling and quasi-real-time processing.
Offline data and real-time processing target different business scenarios, the technical methods behind them are different, and the resource investment is also different. Understanding the positioning differences between them can help to choose the appropriate solution to achieve the enterprise's scheduled business intelligence BI project with the minimum resource investment. The goal of building.