Looking at the past, we found that the development of automobiles and BI has the same purpose.
In the past 100 years, the power of cars has changed from steam to fuel to new energy sources, and the control methods of cars have changed from manual to automatic to intelligent driverless driving. In the field of BI, since the concept of BI was proposed in 1958, the underlying data preparation has ranged from report development, Cube multidimensional model, wide table to indicators; to intelligent augmented analytics. 
Figure: Four Stages of BI Development in China
The comparison found that although automobiles and BI belong to different industries, they have both experienced continuous exploration of technology, and both take intelligence as their future goal to achieve lower threshold, energy saving and efficiency enhancement! As a well-known domestic BI manufacturer, Sematic has been combining industry-leading technology and best practices for more than ten years, and has spared no effort to create good products for users.
On August 8, Sematic released the one-stop ABI platform Smartbi Insight V11 version (hereinafter referred to as V11 version) at the "2023 Smartbi V11 Series New Product Launch Conference" to help enterprises realize digitalization through index-driven and intelligent decision-making operate. Next, we will deeply interpret Smartbi's one-stop ABI platform from three dimensions.
Highlight 1
Index-driven, surging momentum
The engine of a car is like the heart, it can provide power to the car. A good engine running at high speed will make the car run more energy-efficient, fast, stable and safe. The data model and index model of Smartbi Insight's "data management" layer are like the engine of a car. When it is running surgingly, it provides a steady stream of power for the upper-level data application and data consumption, enabling the enterprise to fly on the right road, like the wind and lightning, and run steadily Zhiyuan, safe and orderly!
▎Data model - encapsulation of powerful data processing capabilities
The data model encapsulates multi-source, heterogeneous, and complex raw data. Through convenient dimension index definition, calculation measurement, and member derivation, a data model with strong data processing capabilities, well-organized, and everyone can understand is formed. , so that business users can use data simply and conveniently.
The V11 version further consolidates the data processing capabilities of the data model and lowers the threshold for use. The data model implements dual engines, similar to the "DM-i super hybrid" of a car. The newly added SQL engine can query detailed business data with a large amount of data; in addition, the data model supports wizard-style visual operations to create calculated measures and calculated columns, which is easy to use.
Figure: Wizard-based generation of calculated metrics
▎Indicator model - a bridge for communication of business technical requirements
It is very handy for users to apply data through the data model. However, its construction method is still a bit complicated for business people who do not understand technology, and it is impossible to see the whole picture of enterprise data assets and indicators. Smartbi also provides the indicator model function, through the in-depth business participation and one-stop indicator management service, it can help enterprises build a self-growth indicator system and form enterprise data assets.
The indicator model builds a bridge of demand communication between business users and technical personnel. The two parties use a common interface and language to communicate. Business users use business terms to describe requirements, and technical personnel directly convert it into the indicator model in Smartbi. There is no loss and deformation of information in the process, and the unification of index caliber is realized. At the same time, on the established indicator model, business users can directly derive calculation indicators without the participation of technical personnel.
Figure: Calculation Indicator Derivation
In the V11 version, the indicator model with a unified business and technical caliber can be automatically converted into a data model for subsequent query and analysis with one click. Users can freely select dimensions and indicators according to their needs, and split them into multiple data models. This can not only form a dedicated business model, but also be easy to maintain and have better performance, thereby improving user experience.
Figure: Indicator Model Generates Multiple Data Models
In the V11 version, the indicator model supports direct access to dimension tables and fact tables in the database. When an enterprise has its own data warehouse but lacks a unified indicator system management platform, this function can be used to realize the storage mapping of indicators, so that unified management can be realized without complete reconstruction.
Figure: Dimension table and fact table in the access database
Highlight 2
Free control, intelligent decision-making
The control methods of cars range from manual, automatic to intelligent unmanned driving. The purpose is to reduce the technical difficulty and threshold of driving through high-tech, so that more and more people can enjoy the convenience brought by cars. And our one-stop ABI platform Smartbi Insight provides spreadsheets, self-service analysis tool sets, conversational analysis, data mining and other functions to meet the full scenarios of different BI stages (traditional BI, self-service BI, intelligent BI) in breadth and depth Data application requirements.
▎Self- help analysis—teach a person how to fish, let the business use it
In the V11 version, the content and interface paths of some functions have been reorganized and planned. The new ad hoc query, perspective analysis, and interactive dashboards are more in line with user habits and needs, and are extremely easy to use.
Figure: New impromptu query, self-service detailed data query tool, check and drag to query freely
Figure: Interactive dashboard, with rich components and component templates, visual operation WYSIWYG
▎Conversational analysis - ChatGPT in the field of BI
The conversational analysis function of Smartbi Insight is to conduct data analysis through natural language and data interaction. In the V11 version, conversational analysis uses a new query solution supported by LLM, modeling data through indicators, making it easier for users to understand, providing complex computing capabilities through data models, accurately understanding user intentions through LLM, and finally completing interaction through NL2MQL, providing desired result.
In version V11, conversational analysis supports more ways of use. Independent "conversational analysis" to freely explore data and share results externally; quickly create pages through AI question and answer in the interactive dashboard, and increase the efficiency of large screen & report development by more than 200%; integrate with enterprise IM DingTalk, WeChat, etc., Efficient collaborative office.
Highlight three
New UI, excellent experience
The V11 version uses popular design elements and color schemes to create a new UI style. The overall color is fresh and natural, presenting simple, beautiful, clear and unified visual feedback to users. These well-crafted UI effects can be directly put into use by users.
In addition, we conducted in-depth research on the usage habits of BI users, and re-optimized product modules, interaction methods, and operation methods. Not only the classification is clearer, but also guided operations are provided to improve the usability and efficiency of the product. Users can use it more easily and reduce The cost of user learning and use is reduced, and the experience is more high-quality and enjoyable!
Figure: Add a new navigation page to open up the help system, allowing users to work and study more efficiently
Figure: Launch of "Personal Center" management resources, easy access, organization and sharing of information
In addition to the highlights described above, Smartbi Insight V11 has been enhanced in terms of data source expansion, visual analysis, stability and performance. The V11 version polishes the underlying data preparation, consolidates the upper-level intelligent analysis, and realizes a comprehensive upgrade from the bottom up, thus leading the new navigation mark of digital operation! Welcome to the Experience Center , to "test drive" for yourself!