In 2022, enterprises need to focus on 12 data and trend analysis

Russia and Ukraine caused by long-term geopolitical crisis for the new champions league epidemic raging world is undoubtedly worse.This year data and analysis the leader's job is one of the key management of the resulting continued uncertainty and volatility.

Should now according to the key data and trend analysis technology for the urgency of the business priorities and matching to monitor, attempt, or actively invest in these trends, in order to forecast, adjust and expand the value of the data and analysis strategy.

This year's main data and trend analysis focuses on the following three topics:

• Activate the diversity and vitality.The use of adaptiveAIPromote the growth and innovation, at the same time dealing with the global market volatility.

• Enhance the capacity of staff and decisions,In order to provide the business driven modular components to create rich, situation analysis.

• Trust will be institutionalized,To implement the data and analysis on a large scale value.AI risk management and implementation of edge across a distributed system, environment and the emerging ecosystem management of the Internet.

2022 need to pay attention to the 12 data and trend analysis (D&A)

We have published data and trend analysis represents the business, market and technology developments that nots allow to ignore.These trends will also help determine to promote new growth, efficiency, toughness, and the innovation investment priorities.

Trend: Adaptive AI system (the Adaptive AI systems)

With decision making becomes more relevant, situational and continuity, the importance of reengineering decision-making is increasing.Companies can do this by adaptive AI system, it can more quickly adapt to the change, provide more flexible decision quickly.

At the same time, build and manage the adaptive AI system need to adopt the AI engineering practice.AI工程能够通过编排和优化应用来适应、抵御或吸收各种干扰因素，促进自适应系统的管理。

Download e-book:The combination of human and computer data analysis affect how effective decisions"

Trend 2: data-centric AI (Data - centric AI)

Without considering the AI unique data management problem of trying to solve the problem of AI.Sallam said: "if you don't have the correct data, build the AI can produce risk and potentially dangerous."Therefore, formal rules using a data-centric AI and AI centered data is very important.In enterprise data management strategy, they can be more systematically solve the problem of data deviation, diversity and tags, including active metadata management in data integration and automation data used in the weaving."

Trend 3: Metadata driven data knitting (Metadata - driven data fabric)

Data weaving through the metadata listening, learning and action.Recommended for tags and personnel and system operation, eventually improve the enterprise internal agency to trust and use of the data, reduce 70%, including design, deployment, and operation of all kinds of data management tasks.

Turku, Finland, for example, found that the city's innovation is hampered by the data gaps.By integrating disperse data assets, two-thirds of turku city to use data, reduce product time to market and create a data weave can be converted into cash.

Trend 4: data sharing Always (Always share data)

Although leaders often admit data sharing data and analysis ability is a key of digital transformation, but they lack of professional knowledge, so we can not trust with Shared data on a large scale.

Smoothly push data sharing and increase access to match the correct business case data, should carry out across business and cooperation in the field of industry, it will speed up support for increased investment in budget authority and data sharing.In addition, adopt data weave design should also be considered by the unity of the different types of internal and external data sources data sharing framework.

Trend 5: situation analysis of the rich (the Context - enriched analysis)

Rich situation analysis based on graphics technology.Information about the user's situation and demand are kept in the graph, to take advantage of the relationship between the data points as well as the data point itself more in-depth analysis.This can help you based on similarity, restricting factors, path and the community to identify and create the situation further.

Data to capture, save, and use situation, enterprises need to establish a data line, X-ray analysis technology and AI cloud service ability and skills, in order to deal with different types of data.Driven by the year 2025, situation analysis and the AI will replace 60% based on the traditional data model of existing models.

Trend 6: the Business module combined data and analysis (Business - composed D&A)

Gartner advises enterprises to adopt modular data and analysis method or "combined data and analysis".The business module combined data and analysis based on the trend, the focus is shifted from the IT staff to business people.

The business module combined data and analysis to make business user or by a combination of business and technical personnel should build business-driven data and analysis ability.

Trend 7: data and analysis for the center with decisions (Decision - centric D&A)

Intelligent decision-making subjects (i.e., how to make decisions for thoughtful) is making enterprise organization to rethink investing in the data and analysis ability.Using intelligent subject design best decisions, and then provide the required information and resources.

Gartner predicts that by 2023, more than 33% of the large enterprise institutions will be engaged in the work of the intelligent decision analysts, including decision-making model.

Trend 8: the shortage of the personnel Skills and literacy (Skills and literacy shortfall)

Data and analysis leader needs the talents of the team to push the measurable results.But virtual workplace and contributed to the fierce competition for talent employee data literacy (reading, writing and the ability to transfer data).

From now until 2025, Gartner predicts, most chief data officer (CDO) will not be able to develop a data-driven implementation strategy established business goals required employee data quality.

Due to data quality and employee skills upgrading rising cost of investments should be in the contract with the new employees to join "pay back" or "pay" clause, so you can pay when employees leave.

Nine: Internet governance (Connected the governance)

Enterprise institutions at all levels should take effective governance to solve their current operational challenges, and these measures must also be flexible and extensible and the ability to quickly respond to changing market dynamics and strategic organizational challenges.

Outbreaks, however, further highlights the enterprises urgently need to be strong cross-functional collaboration, and ready to change the organizational structure, in order to realize the agility of business model.

Internet governance should be used to establish a virtual across business functions and geographical data and analysis of governance to achieve the desired cross-enterprise business results.

Ten: AI risk management (AI risk management)

If the enterprise organizations will time and resources to support the AI (TRiSM) trust, risk and safety management, so they can improveAIIn use, business goals, implementation and results about the external and internal user acceptance.

Gartner predicts that by 2026, developed the trustworthy enterprise organizations will achieve more than 75% of goal-oriented AI AI innovation success rate, and institution has not been able to do this is only a 40% success rate.

By strengthening the importance of AI TRiSM, enterprise organization can be controlled and steadily implement AI implementation and operation of the model.In addition, the Gartner also predict the failure of the AI will be reduced sharply, including incomplete AI project, accident or reduce negative results, etc.

Trend 11: manufacturers and regional ecological system (Vendor and region ecosystems)

With regional data security law enacted, many multinational companies is to comply with local regulations and building data and analysis of the ecological system.This trend will accelerate in the new multipolar world.

You will need to consider migrating and copying some or all of the data within a particular region and analysis the stack, and will be cloudy and many manufacturers of strategic management into the design or default.

Enterprises need to take several actions to build a cohesive cloud data of the ecosystem.Extensibility and manufacturers should be evaluated solutions throughout the supply situation of ecosystem, and consider consistent with them.Shall be by a single vendor to weigh the ecosystem in terms of cost, agility and speed advantage, review for the best or most appropriate cloud end-to-end data and analysis ability of strategic policy.

Twelve: trend to the edge of the extension (Expansion to the edge)

In the data center and distributed outside of the public cloud infrastructure equipment, servers, data and analysis activities performed or gateway is increasing.They are more and more in the periphery computing environment, more close to the data and related decision-making place of creation and execution.

Gartner predicts that by 2025, more than 50% of the enterprise key data will be outside the data center or cloud outside the location of the creation and processing.

Enterprise should be the governance capability of the data and analysis to the edge of the environment, and realize visibility through active metadata.But also by joining in IT centered on the edge of the technology (relational and non relational database management system), and is used to store and handle more near the edge of the equipment data and low memory footprint of embedded database, provide support for data persistence of edge environment.