Gartner has celebrated this week in Orlando his event Data & Analytics. The consultant has destined to identify the main trends in data analytics by 2025 and the wide range of challenges which entails these processes, including organizational and human problems.
“The D&A is going from being a domain of a few to be omnipresent”said Gareth Herschel, Vice President of Analysis of Gartner, at the summit where they have analyzed the current and future state of a data management increasingly important in world technology.
«At the same time, D&A leaders are under pressure not to do more with less, but to do much more, and that can be even more challenging because the risks are increasing. There are certain trends that will help leaders to meet the pressures, expectations and demands they face »the executive has stressed.
Data analytics for 2025
The Gartner Data & Analytics Summit 2025 has gathered for three days in Orlando industry leaders, analysts and specialists in data management, governance and architectures, to try to define the main trends for this year that decision makers must address and incorporate into their strategy. The analysis firm He has completed it In the following main sections:
High consumption data products
To take advantage of high consumption data products, D&A leaders must focus on critical use cases for business, correlate and climb the products to relieve data delivery challenges. It is essential to prioritize the delivery of reusable minimum viable data products, which allows equipment to improve them over time. Leaders must also reach a consensus on the key performance indicators between production and consumption equipment, which is vital to measure the success of data products.
Metadata management solutions
Effective metadata management begins with technical metadata and then expands to include commercials to improve the context. By incorporating various types of metadata, organizations can enable data catalogs, data lineage and promoted use cases? It is essential to select tools that facilitate automated discovery and analysis of metadata.
Multimodal data fabric
To develop a solid metadata management practice, it is necessary to capture and analyze metadata throughout the data flow. The knowledge and automation of the data structure support the orchestration demands, improve operational excellence through dataops and enable data products.
Synthetic data
Identifying areas in which data are missing, they are incomplete or expensive to obtain is crucial to advance in AI initiatives. Synthetic data, either as variations of the original data or as confidential data replacements, guarantee the privacy of data and facilitate the development of AI.
ANALYTIC OF AGENTS
The automation of the closed cycle business results with AI agents for data analysis is transformative. Pilot use cases that connect knowledge with natural language interfaces and the evaluation of suppliers’s roadmap for the integration of applications in the digital workplace are recommended. Establishing governance minimizes errors and hallucinations, while evaluating the preparation of the data through the principles of data prepared for AI is essential.
AI agents
AI agents are valuable for ad hoc adaptive automation needs, flexible or complex. Beyond depending solely on large language models (LLM), other forms of analysis and AI are needed. D&A leaders must allow AI agents to access and share data between problems without problems.
Small language models
It is recommended to consider small language models instead of large language models to obtain more precise and appropriate results for the context within specific domains. It is recommended to provide data for the recovery of custom domain models of augmented generation or fine adjustment, especially in local uses to handle confidential data and reduce resources and computer costs.
Compound
The use of multiple techniques of the improvement the impact and reliability of AI. D&A teams must diversify beyond Genai or LLM, incorporating data science, automatic learning, knowledge and optimization graphics for integral solutions.
Decision Intelligence Platforms
It is essential to move from a vision based on data to a vision centered on decisions. It is recommended to prioritize urgent business decisions for modeling, align Decision Intelligence Practices (ID) and evaluate ID platforms. To achieve success, it is essential to rediscover data science techniques and address ethical, legal and compliance with decisions.
Leaders in data analytics can learn more about how to evaluate their own effectiveness using the CDAO de Gartner Effectiveness, an exclusive tool that allows CDA to understand their effectiveness as leaders and discover their strengths and improvement areas.
Gartner analysts will provide additional analysis of the sector in successive peaks that will be held in Sao Paulo, London, Tokyo or Sidney. The consultant offers additional information in its Data & Analytics section.
