AI technology is rapidly updating, affecting human work and life. If you don't pay attention to it or miss this round of opportunities, you may be crushed or eliminated by such a fast-coming & ldquo; Time Train & rdquo. However, as ordinary enterprises and employees, it is not easy to see clearly the essence of this round of technology boom and how to rely on the application of technology to establish a foothold in the market.

Gartner's insights provide people with action guidance.

In the current digital economy era, data has become one of the most important assets of enterprises. Grasping the data is the most effective countermeasure.

Three trends make the pressure of enterprises increase day by day

make good use of the data to grasp the value of the data. We need to have a clear understanding of the data. Gartner believes that data and analysis will present the following three important trends in 2024.

Gartner proposes three important trends of data and analysis in 2024

first, from & ldquo; Pass & rdquo; To & ldquo; Drive business success & rdquo;. According to the survey, 59% of enterprise CEOs believe that artificial intelligence will become the biggest technical driving force to change the industry in the next three years. With the acceleration of technological transformation, people's expectations for technology are also increasing. Most enterprises have initially built a barely usable data analysis platform, now it relies on the power of data analysis and artificial intelligence to drive business success. The challenge lies in how to reshape the value of data analysis and how to apply new data business models such as chain operation of data analysis.

The second is from & ldquo; Chaos & rdquo; To & ldquo; Management complexity & rdquo;. With the change of digitization process and system, there are numerous heterogeneous platforms, isolated data islands and different data analysis, products and capabilities in the field of data analysis and artificial intelligence, and their complexity is increasing. The challenge for enterprises in this regard lies in how to understand the ecology of data analysis so as to better organize these capabilities to cope with the complexity of chaotic situations.

Third, from & ldquo; Overload & rdquo; To & ldquo;AI empowerment & rdquo;. Using AI to improve efficiency and drive business success is the primary goal of enterprise data officers/CIOs. Although the data analysis team is entrusted with an important task, the work pressure is actually getting heavier and heavier. Its countermeasure is to train AI-ready employees and possess AI-ready data.

Many people regard AI as a key supporting tool. However, the survey shows that more than half of the enterprises pay more attention to & ldquo; Produce more data products & rdquo; And & ldquo; Application empowerment business & rdquo;, rather than & ldquo;AI & rdquo; Or & ldquo; Application and implementation of generative AI & rdquo;.

This verifies the real demand of enterprises for the core ability of creating business value, and is also the driving force for enterprise data officers/CIOs to drive business success and prove their own value with data. Gartner predicts that 75% of these people may be re-rotated or replaced in 2026 if the influence of data is not taken as the top priority.

The more severe reality is that most enterprises are not ready for data: 55% of enterprises think it is difficult to achieve data AI Readiness, 37% think it is expected to achieve data AI Readiness, only 4% of enterprises believe that they have achieved Data AI Readiness.

A data AI ready tool has three key points: first, it has a high-quality data analysis and governance platform, which can establish standards for data quality, data lineage and other aspects, to ensure that high-quality models can be obtained in a high-quality environment; Because different data require different processing methods in different situations, how to understand these data or match them with optimal scenarios based on these tags and metadata means the key value of metadata management in AI Readiness; Data is constantly changing, its changes will directly affect the establishment and application of the model. How to Observe potential anomalies through unknown data in an unknown environment and identify whether AI models are applicable or may have negative effects in advance are the most critical factors in AI application. Observability of data is crucial.

AI Readiness cannot be separated from the data base.

Both AI and powerful models cannot be separated from solid, accurate and reliable data sources. Only by doing a lot of compression, analysis and processing based on them can artificial intelligence applications be generated.

Gartner predicts that by 2026, the expenditure on technology and services for unstructured and semi-structured data will account for 40% of the total expenditure on enterprise data management, while now this proportion only accounts for 5%. In the future, more enterprises will enhance their awareness of the importance of data and regard data as the core element. This coincides with the country's advocacy and emphasis on the application of data assets.

AI technology and data processing also require a large number of new methods to help. The advantage of human beings lies in a large amount of innate knowledge and experience, filling in the details of AI deficiency. The combination and complementation of the two will better fulfill the business value.

Understand the actual business needs and let AI give wings to data analysis.

The data analysis required by the business is actually very simple: it can be directly combined with daily business, increase sales revenue, reduce customer churn and a series of quantitative business results, rather than fancy dashboards, tools such as self-analysis reports.

For data analysts, the most important thing is to be able to go to the front line of the business, understand the actual needs of the business and achieve the above goals through appropriate tools.

Gartner also provides a specific framework. A classic solution is & ldquo;Gartner decision intelligence model & rdquo;, which can help enterprises match and summarize data analysis capabilities with core business processes, help enterprises to make key decisions better and ensure that business value can be realized in the whole business chain.

To drive business success, you need to select an appropriate operation mode. In response to different regions, different data maturity levels and regulatory environments, Gartner proposed a model named & ldquo; Franchise & rdquo; And believed that, 60% of multinational enterprises will adopt this model by 2027.

& ldquo; This model is similar to & lsquo; Franchise & rsquo; Which is set up in KFC. It is also applicable to the team and architecture of data analysis. & rdquo; According to Fang Qi, senior research director of Gartner, this model develops corresponding processes and capabilities and customizes products and services through different regional characteristics and different business objectives, the integration and collaboration of back-end processes, on the one hand, ensures its stability through efficient governance, on the other hand, better responds to the unique needs of different businesses through independent and agile innovation capabilities.

Fang Qi, senior research director, Gartner

it can be seen that the & ldquo; Franchise & rdquo; Model allows the enterprise data architecture, organizational structure and governance architecture to have both & ldquo; Freedom & rdquo; And & ldquo; Control & rdquo; the advantages of the two aspects, and achieve a dynamic balance between the two.

Gartner believes that the success of the support & ldquo; Franchise & rdquo; Model cannot be separated from three key factors.

One is to build a centralized and decentralized organizational structure.

Use the capabilities of & ldquo; Centralization & rdquo; To unify and control information architecture, data architecture and data governance. At the same time, use the capabilities of decentralization to support finance, human resources, supply chain, decentralized organizations and functional centers such as procurement and marketing innovate in professional fields.

The second is to form a cross-functional team.

Gartner predicts that 90% of data analysis users will use AI to create content by 2025. This means that employees who do not apply AI will definitely be eliminated by employees who are good at AI applications. More than 1/3 of enterprises believe that business-side data analysts need AI empowerment most. However, IT is difficult for a single organization to meet the extensive needs of knowledge fields such as IT, Data scientific analysis and artificial intelligence at the same time. Therefore, a cross-functional team is formed to flexibly, the way of integration to achieve the same goal is the key element in & ldquo; Franchise & rdquo.

In Gartner's view, the most important thing for employees to have AI literacy is to have basic knowledge, ability and understanding of risk value. They realize that AI is not a simple substitute for people, AI also has its advantages and disadvantages. Instead of blindly following AI, it applies AI in the context to optimize and change the process so that the entire business value can play a greater role. For the development of AI or the construction of AI, it requires more comprehensive capabilities, which can be grasped from the aspects of value, basic engineering and governance, so as to better apply corresponding technologies to solve business problems, drive change and better enlarge domain knowledge. The third level of AI literacy is the ability to cooperate across domains, with a deep understanding of business, digital and data literacy and artificial intelligence, can integrate different qualities to develop and apply AI.

The third is the governance framework with prototype as the priority. Gartner proposed this framework theory because enterprises are facing rapid business changes and it is difficult to capture all business requirements in one step at the initial stage of planning data analysis and AI product capabilities. Therefore, Gartner proposes to build some small-scale application prototypes that can be quickly iterated for deployment. After short-term application and trial-and-error verification, Gartner continuously summarizes and updates them, after maturity, it will be widely promoted.

The development and application of AI and big models are moving towards rationality.

Gartner's technology maturity curve shows that the current generated AI and big models are undoubtedly the & ldquo; Peak & rdquo; Mentioned in many reports last year, which means that the market has high expectations for related technologies, but in fact, there will be great differences between technology and maturity. In fact, a large number of attempts on large models or generative AI in the past have made people clearly perceive such a conclusion. Gartner noted that enterprises have more rational perceptions of the actual business value that AI can produce, and eventually related technologies will inevitably fall from the peak and then go through the trough, slowly climb to & ldquo; Mature period & rdquo;.

Fang Qi said that the generation AI technology will be promoted by the whole market one after another and will eventually become mature. However, some of these technologies inevitably go to death and & ldquo; The Valley of calmness & rdquo;.

Gartner guides enterprises to drive business through technology

the survey found that 59% of enterprises believe that the primary obstacle to deploying data analysis and artificial intelligence is how to integrate various data sources from heterogeneous and complex systems.

Gartner proposed many years ago that it can flexibly deploy different applications such as data warehouse, data integration, database, data Lake, artificial intelligence, and business intelligence through data assembly, meeting business needs also requires a more thorough understanding of enterprise needs and timely updating of these core capabilities so that technologies can be reused more.

Gartner provides the most classic help for enterprises to cope with reality: & ldquo; Magic Quadrant & rdquo;. For data analysis and AI AI platforms, Gartner uses a large number of & ldquo; Magic Quadrants & rdquo; Reports to evaluate manufacturers' capabilities horizontally and vertically. This series of reports derived from market data have become the guidance and specific suggestions of enterprise customers.

Gartner pointed out that in the uncertain world, the only certainty for an enterprise is to drive its business through AI and other technologies, change its business, facilitate its business and transform its business through data and analysis, and make the business gain value.

END

on August 28th, the 2024 global Flash Summit will be held grandly in Nanjing. Six thematic forums will be presented throughout the day, bringing together nearly 100 experts from industry, university and research institutes to communicate with you on the spot to interpret new Flash technologies and new trends of enterprise-level applications, release the Flash industry database and Panorama, flash cloud list, a overview of the development history of flash memory, more exciting content, please look forward to it!

Source: DOIT media