Data Modeling, the practice of diagramming Data Architecture and aligning core business rules with data definitions, will go through a refresh in 2023. Organizations will assess and make over their data fabrics, which describe distributed Data Management systems. Furthermore, managers and project teams will use data models to optimize business processes with their data assets and allocate resources to achieve these goals.
IDC reports that 60% of organizations want to enhance products, services, and customer experiences digitally. Additionally, companies plan on increasing spending on digital activities. Yet, firms will be resourceful in allocating their IT spending to modeling as business costs continue to rise. So, Data Modeling will focus on high-impact activities to squeeze out more benefits with fewer costs. To these ends, Data Modeling will center around the following seven trends in 2023:
- Leveraging AI engineering
- Identifying changes to existing data models
- Doing model-driven development
- Modeling big data
- Adopting small and wide data models
- An increasing role in Data Governance, which formally manages data assets
- Investing in flexible and scalable Data Modeling tools
Leveraging AI Engineering
Artificial intelligence (AI) engineering will take a more critical role in remaking Data Modeling by handling more of its operations and providing insights into data flow and mapping. By taking the mundane and cumbersome modeling tasks, AI will free data modelers to do more business-centric modeling. As a result, data modelers will more quickly create relevant definitions, relationships, and conceptions of business issues and opportunities.
AI improvements in revising outdated data models and suggesting changes to data models quicker will get a pulse on customer engagement. Furthermore, they will figure out why and with better insights than the competition.
Data Modeling for AI systems will also improve, into 2023, through adaptive AI. AI systems will evolve to operationalize better their modeling approaches to producing insights.
Adaptive AI engineering will offer additional insights into efficient data movement, flexibly adjusting data models in response to blockages and openness in data pipelines. Firms will leverage as much AI engineering as they can to facilitate their modeling goals.
Identifying Changes to Existing Data Models
Identifying and understanding how Data Modeling changes through time series databases will increase in prominence in 2023. Time series databases help analysts and data modelers identify patterns in data representations to understand the information context. As a result, data modelers update their diagrams according to time windows to boost data flow and processing.
For example, the United States Census Bureau recognizes necessary seasonal adjustments to its monthly and quarterly data. Additionally, to get a handle on how a data model needs to change for better analysis, the Census Bureau plans on using daily and weekly time series databases. These tools will better filter and forecast data changes. The Census Bureau plans on building on this Time Series and Seasonal initiative into 2023 and beyond.
As 2023 gets underway, companies will increasingly leverage time series databases to do more real-time modeling. Also, AI-driven capabilities will use time series to recommend how data models need to evolve with business contexts.
Doing Model-Driven Development
Data Modeling activities will play an increasing role during product creation and enhancement. Consequently, more companies will use model-driven development, a process of modeling and remodeling data alongside the actual code and data deliverables.
Model-driven development works well with companies’ iterative processes, focusing on rapid development. Data Modeling will adapt as software platforms mature and take on living and growing data viewpoints.
Data Modeling reviews and updates will occur and evolve with product feature requests. As part of changing data diagrams, modeling components will get reused, encouraging a modular approach when creating data models. This trend follows Gartner’s prediction that modular constructions to composable data and analytics will grow in 2023 and beyond.
Modeling Big Data
Larger companies with more abundant resources will refresh their big data models, connecting disparate systems more effectively. Thomas C. Redman notes that companies must consider their entire big data infrastructures, working above, around, and through siloed information.
Consequently, data modelers will expand existing diagrams, showing data flow from ingestion to consumption. Data Modeling will transform, playing a predominant role in defining data supply chains.
These repurposed big data models will describe data processes between existing data lakes, data warehouses, cloud computing, and other systems. Big data models will look like factory floors where data moves from one packaging station to another.
Big Data Modeling will shift focus from a database design to data ecosystem construction. Given the massive undertaking to model big data, companies with fewer resources will consume Data as a Service (DaaS) products from larger and more established firms.
Adopting Small and Wide Data Models
Many firms, in 2023, will not have the luxury to do Big Data Modeling. They will turn to small and wide data models, a powerful alternative requiring less data but providing more insights. According to Ravi Shankar, chief marketing officer at Denodo:
“Organizations will leverage small data analytics to create hyper-personalized experiences for their individual customers to understand customer sentiment around a specific product or service within a short time window.”
With small and wide modeling, businesses focus on extending existing capabilities and data models. Use cases include demand forecasting in retail, customer experience improvement, physical security, and fraud detection. Small and wide data modelers take existing diagrams and refresh their data components for the new goal.
Gartner predicts that 70% of companies will shift their focus from big data to small and wide data by 2025. This upward trend towards adopting small and wide data comes as no surprise. Only 15.54% of our aforementioned 2022 data management survey’s participants implemented big data ecosystem technologies.
An Increasing Role in Data Governance
Data Modeling will increase its role among Data Governance teams throughout 2023. As organizations cobble siloed information together and smooth out disagreements about different Data Management approaches, Data Governance will turn to data models as an efficient means.
Data modelers will ground existing generalities and abstract conceptions about organizational data. For example, a Data Governance team can see which data store components present the best choice for analyzing and updating existing physical models.
Once a Data Governance council agrees on describing core data concepts and definitions, they can use modeling to track their alignment. Then the Data Governance team will further refresh data models, tackling other data definitions or relationships or guiding service development to best impact the business.
Data Modeling Tools
Finding the best modeling tools in 2023 will increase in importance as companies figure out how to optimize business processes. This trend will encourage companies to eliminate modeling tools that focus on older technologies for those that serve newer objectives.
Data Modeling tools that leverage AI and automate routine tasks will look attractive to firms. However, one tool that uses AI for modeling may not represent a good choice to do small and wide data vs. big data. Choosing a modeling tool will depend heavily on an organization’s needs.
In 2023, Data Modeling will describe multiple organization platforms and systems, making up a data fabric. This kind of Data Modeling will refresh older approaches to visualizing data assets.
Image used under license from Shutterstock.com