What’s in store for the world of Data and how does it continue to rapidly evolve for organisations? Siarhei Oshyn, Head of Data shares his reflections and predictions on how Data will evolve in 2025.

How has the role of data analytics have evolved over the past year?

Over the past year, the role of data analytics has undergone a significant transformation, driven by the growing need for actionable insights, advanced technologies, and the convergence of AI with analytics. Companies across industries are investing in innovative tools and strategies to leverage data more effectively. Below, we explore the key trends shaping the data analytics landscape.

The Continued Expansion of Data Warehousing

Organisations are increasingly focused on building robust data warehouses to extract valuable insights from their growing volumes of data. Modern tech stacks, including Amazon Redshift and Amazon QuickSight, Azure Synapse and Power BI, Snowflake and ThoughtSpot, empower businesses to transform raw data into accessible, actionable intelligence. These tools allow users to seamlessly visualise and explore data, fostering data-driven decision-making across teams.

Data warehouses are no longer just storage systems; they’ve become the cornerstone for business insights, enabling companies to maintain a competitive edge.

Moving Beyond Descriptive Analytics

Traditional reports answering “what happened?” are now considered essential in business operations. However, the demand is shifting toward predictive and prescriptive analytics, which answer critical questions like “What should we do next?”

This shift is fuelled by advancements in AI-driven analytics. Solutions like AutoML, scenario-based forecasting, and dashboards with embedded actionable recommendations are empowering decision-makers to stay proactive. By integrating these AI capabilities, businesses can forecast trends, evaluate potential outcomes, and devise data-informed strategies, turning analytics into a forward-looking, strategic advantage.

Embracing Generative AI

Companies are increasingly exploring the potential of Generative AI to redefine customer interactions and operational efficiencies. By leveraging AI chatbots built on GenAI models like Mistral AI or Llama, organisations can harness the power of conversational AI to deliver personalised experiences, streamline customer service, and enable seamless query resolution.

Additionally, businesses are developing Knowledge Bases, integrating internal and public documents into centralised repositories. These systems provide a foundation for AI-powered insights, ensuring knowledge is readily accessible for both customers and employees.

Innovative Approaches to Text Extraction with Large Language Models

Traditional ETL processes for extracting text and structured data from documents are being replaced by the capabilities of Large Language Models (LLMs). These models enable businesses to analyze and extract insights from complex, unstructured data sources with unprecedented accuracy and speed.

This shift not only reduces dependency on traditional data pipelines but also opens new possibilities for processing diverse document types, from contracts to invoices, while ensuring scalability and efficiency.

Text-to-SQL AI Chatbots

One of the most exciting advancements is the adoption of AI chatbots capable of generating SQL queries based on natural language inputs. Tools like Copilot for PowerBI or Amazon Q for QuickSight, or Semantic Kernel enable business users, regardless of technical expertise, to create custom reports by simply asking questions.

For example, a user might request, “Show me the monthly sales growth for the past year,” and the chatbot instantly generates a report based on the table data in the database. This innovation bridges the gap between technical complexity and business needs, fostering accessibility and agility in data-driven decision-making.

How will the continued growth of big data affect data storage and processing infrastructure in the next 5 years?

Scalable and Cost-Effective Storage Solutions

  • Cloud Storage Dominance: As data volumes grow exponentially, more organisations will turn to scalable cloud storage platforms (AWS, Azure, Google Cloud). Pay-as-you-go models will allow businesses to handle fluctuating data needs.
  • Edge Storage: With the rise of IoT and real-time data needs, edge computing will grow, necessitating localised storage solutions closer to data sources to reduce latency and bandwidth costs.

Advances in Data Processing Infrastructure

  • Distributed Computing: Frameworks like Apache Iceberg and Apache Spark will continue to evolve to handle larger datasets efficiently across distributed systems.
  • Real-Time Processing: As the demand for instant insights grows, platforms such as Apache Kafka and Flink will see increased adoption for stream processing.
  • AI and ML Integration: Big data processing infrastructures will be optimised for AI/ML workloads, incorporating GPUs, TPUs, and specialised processors.

Enhanced Data Management and Governance

  • Data Fabric Architecture: Integrated architectures that unify data management across environments (on-premises, cloud, and edge) will become critical for simplifying access and governance.
  • Automation and Orchestration: Automation tools will streamline data integration, backup, and lifecycle management, reducing manual intervention and errors.
  • Compliance-Driven Storage: Regulations like GDPR, CCPA, and others will influence storage solutions to provide secure, compliant, and auditable data management practices.

New Storage Media and Technologies

  • NVMe and SSDs: Faster, high-capacity solid-state drives and NVMe (Non-Volatile Memory Express) storage will replace older spinning disk technologies for performance-intensive applications.
  • Quantum Storage: Advances in quantum computing could pave the way for new paradigms in storage and processing, though practical applications may be limited in this time frame.

Security and Privacy Enhancements

  • Data Encryption: End-to-end encryption and homomorphic encryption will become standard practices for protecting sensitive data.
  • AI in Security: AI-driven security measures will help detect and mitigate threats in real-time, protecting big data infrastructure from cyberattacks.

How will the combination of AI and edge computing enable faster, more intelligent data processing?

The combination of AI and Edge Computing

Edge computing involves processing data near its source, rather than sending it to a centralised cloud server. AI models at the edge allow systems to analyse data in real-time, unlocking several benefits:

  • Low Latency: Local processing reduces the time for data round-trips to the cloud, which is essential for time-sensitive applications like autonomous vehicles.
  • Bandwidth Optimisation: Only critical insights are transmitted to the cloud, conserving network bandwidth.
  • Enhanced Privacy: Sensitive data can be analysed locally, reducing the risk of exposure during transmission.
  • Resilience to Connectivity Issues: Edge devices can operate independently of the internet, ensuring continuous functionality in remote or mobile scenarios.
  • Real-Time Decision-Making: AI at the edge enables immediate responses in applications such as industrial automation and healthcare diagnostics.

Powering Edge AI Solutions

Amazon Web Services (AWS) provides a comprehensive suite of tools for deploying AI and edge computing, empowering businesses to process data efficiently and securely at the edge.

  • AWS IoT Greengrass: Enables AI inference, data processing, and messaging on edge devices and facilitates local analysis for IoT applications, like smart factories and connected vehicles.
  • Amazon SageMaker Edge Manager: Helps deploy, monitor, and update machine learning models on edge devices.
  • AWS Panorama: A computer vision service that uses edge AI to analyse video streams from existing cameras. Applications include monitoring inventory in retail stores or detecting safety issues on factory floors and are ideal for retail stores running recommendation algorithms locally to personalise promotions.

Microsoft Azure offers versatile solutions to extend cloud intelligence to edge environments, enabling organisations to innovate with AI and data processing close to the source.

  • Azure IoT Edge: Brings cloud services and AI workloads to edge devices and supports containerised deployments for flexible and efficient operations.
  • Azure Percept: A complete platform combining hardware and software for AI at the edge and Includes vision and audio kits for applications like object detection and voice recognition.
  • Azure Machine Learning (Azure ML) Provides tools for training, deploying, and managing machine learning models at the edge and supports predictive maintenance and smart city applications.

What are your tech predictions for 2025?

AI Everywhere: Democratising Intelligence

Artificial Intelligence will continue to dominate the tech landscape, moving from specialised use cases to universal adoption. Tools that simplify AI model creation—requiring little to no expertise in machine learning—will become increasingly popular.

Low-code and no-code platforms will empower more teams to build and deploy AI solutions, transforming AI from a domain of specialists to an everyday tool accessible across organisations. This democratisation will enable businesses to rapidly scale AI initiatives without the steep learning curve traditionally associated with the technology.

One major development will be the rise of “co-pilot” tools—AI-powered assistants that help professionals write code, automate workflows, and accelerate tasks, fundamentally altering the way teams operate.

From Experimentation to Production

The trend of moving AI and ML projects from experimental phases into production environments will accelerate. Companies will prioritise integrating AI directly into their operational pipelines to extract real, measurable value.

Platforms such as Kubeflow, AWS SageMaker, and Azure ML will see widespread adoption as organisations seek scalable, reliable environments for model development, deployment, and monitoring. These platforms will provide the infrastructure needed to bridge the gap between innovation and practical application.

The Rise of AIOps Engineers

As AI becomes a core component of business operations, the role of MLOps engineers will evolve into AIOps engineers. These professionals will focus on managing the lifecycle of AI systems, ensuring performance, scalability, and cost-efficiency while also addressing ethical considerations and compliance.

AIOps engineers will play a critical role in automating the monitoring and maintenance of AI-driven systems, enabling businesses to focus on innovation rather than operational overhead.

Cost-Efficient AI Solutions on the Rise

The growing emphasis on cost-effective AI solutions will shape the market. Businesses will seek to optimise their AI investments, balancing performance and affordability. This trend will drive innovation in tools and techniques that reduce the computational and financial burdens of AI, from efficient neural network architectures to energy-conscious processing frameworks.

Smarter resource allocation and cloud-based AI services will also help organisations achieve cost efficiencies without sacrificing capabilities.

The Dawn of Quantum Computing

While still in its early stages, quantum computing (Amazon Braket or Azure Quantum) will begin to make a tangible impact by 2025. For fintech and insurance industries, this technology will unlock new possibilities in niche applications such as portfolio optimisation, fraud detection, and complex risk assessment.

Quantum computing’s ability to process vast amounts of data in parallel will enable breakthroughs in solving problems that are currently computationally prohibitive. As early adopters explore its potential, quantum computing will move closer to mainstream adoption, setting the stage for transformative changes in data analytics and AI.