Industry executives and experts share their predictions for 2023. Read them in this 15th annual VMblog.com series exclusive.
2023: The Year of Data and Analytics for Everyone
By David Mariani, Co-Founder and CTO, AtScale
Over the course of the past year, we have seen the enterprise data and analytics market continue to grow and evolve. Incorporating strong data analytics into your organization’s decision-making process has finally moved from a “should we” to a “how do we” discussion. We’ve seen some promising trends this year that really help to drive more data democratization.
In 2023, I anticipate that a few of these important trends will become commonplace and the de rigueur for delivering data and organizing analytics teams at scale.
The Semantic Layer in the Modern Data Stack
The idea of having a semantic layer to support self-service data and analytics has become even more relevant – and we’ve been able to watch the term (and its little cousin the “metrics layer”) spread across the industry. Because of the strong, continued focus on promoting enterprise-wide data and analytics success, I feel strongly that 2023 will be the year that the semantic layer itself becomes recognized as an independent technology category within the modern data stack.
The historically siloed nature of data within an organization has made it difficult for organizations to actively empower employees to utilize data. Companies keep data in different data formats, locations and data platforms and use multiple tools to access, prepare and consume their data.
As a consequence, governance nightmares, conflicting outputs and stale data have become the norm.
The semantic layer combines the capabilities of metrics layers, data modeling, and workflow orchestration with an open integration with data and analytics governance solutions. The semantic layer helps to bring all of the disparate data back together into one place, where conflicts are resolved and a logical, business-friendly view of that data can make data actionable and accessible by more people in the organization.
The importance of this “layer” will drive its relevance as a core component in the modern enterprise data stack. As an independent technology category, the rise of the semantic layer will also lead the industry to another prediction: next year will see the job requirements of analytics engineers evolve.
The Advent of the Analytics Engineer
The enterprise analytics engineer will be counted on by organizations to be infrastructure-savvy, business-oriented technicians that will manage pipelines and data products (by leveraging the semantic layer) for data consumers. The requirement to deliver faster and more accessible data to more decision-makers will drive the morphing of the traditional data engineer, BI engineer and data modeler personas into a single, more valuable analytics engineer that will drive the use of data to improve decision-making across the organization.
Data Science and Business Intelligence Converge
With the semantic layer as the glue, we will also see the rise of elite business analysts that can consume and use AI byproducts – like predictions and feature insights – with the same ease that they use historical data. We will see AI model deployment rates improve as business users infuse AI intelligence in their decision making processes by breaking down the traditional silos separating data science teams from business intelligence teams.
The Rise of the “Hub and Spoke” Analytics Organization
As these trends converge, in 2023 we will also see more organizations adopt a decentralized style of delivering analytics and data to their users. They will accomplish this by leveraging and adapting the principles of data mesh to create a “hub and spoke” approach for delivering data products, incorporating a semantic layer for sharing data models across business domains.
A “hub-and-spoke” approach combines the best of both worlds, while delivering on the promise of data mesh. With this approach, a central data team or “center of excellence” owns the data platform, tooling, and process standards, while business-embedded data stewards own data models for their respective business domains.
A semantic layer operationalizes this emerging approach with a domain-oriented, self-service design that allows domain owners to leverage a semantic model for communicating and sharing data products with their peers. This approach will improve the efficiency and effectiveness of data and business teams and ensure that decisions are made based on real-time, consistent business data.
Anyone can be an Analyst
As the semantic layer becomes increasingly ubiquitous in the new year, organization-wide data literacy will improve and enable all stakeholders to make better-informed, data-driven decisions.
With a semantic layer, data consumers will instantly apply insights without needing training or the expertise to understand the physical structure of the data itself. Organizations can become more prepared and proactive to compete in their ever-changing business environments.
By fostering a culture of self-service and data literacy, anyone and everyone can use analytics to make data-driven decisions.
##
ABOUT THE AUTHOR
Dave is one of the co-founders of AtScale and is the Chief Technology Officer. Prior to AtScale, VP of Engineering at Yahoo!, where he built the world’s largest multi-dimensional cube for BI.
