We already know comms pros are skeptical of the data they work with, and the data managers who provide it, and now new research from data integrity firm Precisely, in collaboration with Business Application Research Centre (BARC), finds a more systemic issue.
The newly released Future of Data Architecture study reveals major discrepancies in how data architecture and management is viewed across organizations—with central data and analytics teams generally satisfied with the future viability of their data landscapes, but business users unconvinced that existing architecture is modern enough to meet constantly evolving business needs.
The report also shows that while 70 percent of business users believe that implemented data and analytics applications do not cover their current or future requirements, only 32 percent of central data and analytics teams share the same sentiment. Furthermore, 52 percent of business users agree that existing data is not suitable for necessary business analysis, compared to only 36 percent of central data and analytics teams. The results show a clear disconnect between data and analytics teams and their key stakeholders—with business users unable to rely on trusted data for confident decision-making.
Stakeholders think the analysis tools are not the limiting factor, but rather the data landscape:
“As businesses increasingly focus on becoming more data-driven, they are being faced with huge volumes of data being generated at accelerated rates,” said Emily Washington, senior vice president—product management at Precisely, in a news release. “The report clearly indicates a growing need for organizations to ensure that existing data management and architecture can meet, and adapt to, the fast-changing needs of the modern business. If organizations are lacking a clear strategy for data integrity, they will quickly find that the data fueling their most important decisions is inaccurate, inconsistent, and lacking vital context.”
The research further reveals the key challenges highlighted by business managers, including extensibility of real or near-time requirements (41 percent), comprehensibility of the data landscape or architecture as a whole (39 percent), and flexibility of extended data requirements (32 percent).
Data culture does not only concern data consumers:
The main drivers for modernization are the optimization of existing data models and processes, migration to cloud platforms, and efforts to improve the quality of source data or data interfaces.
Despite facing challenges with rising volumes of disparate data sources, the research revealed that 50 percent of organizations are continuing to rely on data warehouses as a design paradigm when redesigning their data architectures. However, executives and business users criticize the fact that centralized approaches cannot prevent the emergence of further data silos. Therefore, companies are looking to improve their source data to streamline their data pipelines and build a better basis for data virtualization.
“It’s encouraging to see that business leaders are increasingly focused on breaking down data silos, improving the quality of their data, and enriching it for maximum context. To achieve these goals, organizations must implement a robust strategy for data integrity—combining data integration, data governance and quality, location intelligence and data enrichment to provide a foundation of trusted data that can be shared across the business,” said Washington. “Whether the organization needs to automate decision-making, move fast and reduce costs, or manage risk and comply with complex regulations—a modern, and future-proof, data architecture is crucial to remaining agile in challenging times.”
Most modernization efforts follow old concepts:
Addressing data integrity disconnects
The utilization of ETL (Extract, Transform, Load) and ELT (Extract, Load, Transform) processes plays a pivotal role in navigating the data integrity disconnects between data teams and business stakeholders. ETL typically involves extracting data from various sources, transforming it into a consistent format, and then loading it into a destination system. This method ensures data quality and consistency but can be time-intensive, which may lead to delays in data availability for stakeholders.
On the other hand, ELT reverses the transformation and loading steps, allowing for quicker data ingestion into the target system. While ELT facilitates faster access to raw data, it sometimes compromises data quality checks and consistency, which are critical for accurate decision-making by stakeholders.
Bridging this gap requires a balanced approach: educating stakeholders on the complexities and trade-offs of ETL versus ELT, involving both teams in data integration planning, and fostering transparent communication about data quality expectations and processing timelines.
Download the full report here.
The study was based on the findings of a world-wide online survey conducted in March and April 2022. 268 responses were analyzed for the report, with global representation as follows: Europe (52 percent), North America (35 percent), Asia and Pacific (8 percent), South America (3 percent) and Africa (2 percent). The target audience was primarily data and analytics leaders or executive/C-Level roles, with representation across several industries—including IT, banking and finance, public sector, and retail.
