Trends in Designing Data Analytics and BI Products in the Era of GenAI
Julian, Joy·
As a loyal reader and fan of Gartner, I've always admired the profound insights and expertise their analysts bring to the table. These top-tier experts, with years of industry experience, excel in independent thinking across academic, technical, product, industry, and business domains. Their analyses and reports, enriched by real-world practice and frequent interactions with clients and CXOs, are known for their sharp insights and accurate trend predictions.
For instance, back in June 2022, Gartner accurately forecasted the top trends for GenAI applications in the industry.
In this article, I aim to delve into how the design trends for data analytics and Business Intelligence (BI) products are being reshaped by the rise of GenAI. Drawing upon Gartner's extensive research and deep understanding of the data analytics and BI landscape, I aspire to share actionable insights and practical guidance to help professionals navigate this transformative era.
When it comes to this direction, Gartner categorizes the key products and technologies into two major fields: Analytics and Business Intelligence (ABI) and DSML (Data Science and Machine Learning). Let's dive deep into each.
Analytics and Business Intelligence (ABI) is a broad term that encompasses applications, infrastructure, tools, and best practices, all designed to enable users to access and analyze information to improve and optimize decision-making and performance.
Traditional drag-and-drop interfaces in ABI platforms are now complemented by conversational, text-based interfaces powered by GenAI, allowing users to request reports or pipelines in natural language.
Major cloud ERP and CRM vendors influence ABI platform choices, raising concerns about vendor lock-in, while driving adoption of "multi-cloud" strategies for flexibility and openness. Microsoft leads the market with Power BI's growth driven by its affordable bundling with Microsoft 365 (E5) and its integration with Microsoft Teams, especially in the era of remote work.
Niche analytics vendors differentiate by offering tailored solutions for specific market segments, emphasizing independence from large cloud providers to address lock-in concerns.
The market is rapidly adopting low-code/no-code capabilities, evolving from traditional dashboards to delivering contextual insights that enhance decision-making and drive business value.
Data Science and Machine Learning (DSML) platforms provide end-to-end support for the full lifecycle of AI models, including GenAI. They bridge the gap between development and production, enabling organizations to strengthen MLOps practices and streamline AI deployment.
DSML platforms are increasingly vital as enterprise assets, with surging demand for GenAI driving significant spending growth. However, integrating data, models, and infrastructure into scalable solutions remains complex.
DSML platforms have evolved into full-stack solutions, covering multi-cloud infrastructure, data pipelines, model training, deployment, and front-end development. Differentiation lies in their abstraction levels, enabling rapid iteration without requiring technical deep dives.
GenAI accelerates the democratization of data science, empowering business-aligned roles through mature AutoML features like coding assistants, natural language queries, and workflow automation.
While major cloud providers dominate DSML adoption due to robust infrastructure, opportunities remain for independent players to innovate, particularly in fostering team collaboration.
Amid the GenAI wave, foundational data science use cases that drive actionable, insight-led decision-making must not be overlooked. DSML platforms are uniquely suited to unify advanced analytics and AI development.
By 2026, 50% of organizations will evaluate ABI and DSML platforms as unified solutions due to market convergence. The integration of ABI and DSML is becoming a significant trend, driving a holistic approach to analytics and machine learning.
When designing a product, it is crucial to consider two key aspects:
Product Capabilities
These are the essential features and functionalities a product must have to operate effectively within its domain.
Use Cases
The primary user personas and their application scenarios, each with unique priorities and varying emphasis on specific product capabilities.
Mapping use cases to the corresponding product capabilities is central to guiding product design, ensuring the platform meets user needs while aligning with core objectives.
The ABI product capabilities can be classified into 12 categories:
Analytics Catalog: Enables the display of analytical content, making it easy for users to discover and utilize resources. Supports search functionality and provides recommendations.
Automated Insights: Leverages ML to automatically generate insights, such as identifying the most important attributes in a dataset.
Collaboration: Facilitates teamwork by integrating collaboration into analytical workflows, allowing a wide range of users to work together on projects.
Composability: Offers low-code and no-code tools (e.g., APIs/SDKs) to build modular, flexible user interfaces and embed analytics into workflows. Often integrates GenAI for enhanced functionality.
Data Preparation: Supports drag-and-drop operations, user-driven data source combinations, and the creation of analytical models like custom metrics, collections, groupings, and hierarchies.
Data Science Integration: Enhances prototyping and development, enabling data scientists to create composable ML models and integrate them with broader ecosystems.
Data Storytelling: Combines interactive visualizations with narrative techniques to deliver insights in a compelling, easy-to-understand format for decision-makers.
Data Visualization: Provides interactive dashboards and data exploration through various visualizations, such as heatmaps, treemaps, geographic maps, scatter plots, and more.
Governance: Tracks data usage and manages how information is shared and promoted to ensure quality, compliance, and control.
Metrics Layer: Offers a virtualization layer to define metrics as reusable assets, manage them from data warehouses, and support downstream analytics, data science, and business applications. Includes goal management functionality.
NLQ: Enables users to ask questions about data using text or voice commands, simplifying interaction with analytics.
Reporting: Delivers pixel-perfect, paginated reports that can be scheduled and distributed to large user groups for consistent and reliable insight delivery.
Here're four main ABI use cases: Analytics Developer, Business Analyst, Augmented Consumer, Data Scientist.
Analytics developers are professionals within data teams who are responsible for creating and distributing analytical content to a large user base across the organization.
Critical capabilities in this use case:
Metrics Layer
Composability
Governance
Reporting
Data Visualization
Analytics Catalog
Collaboration
Natural Language Query
Automated insights, data science integration, and data storytelling are less relevant in this use case.
Business analysts refer to professionals who integrate various data sources for visual analysis with minimal reliance on IT departments.
Critical capabilities in this use case :
Data Visualization
Automated Insights
Data Preparation
Analytics Catalog
Data Storytelling
Metrics Layer
Collaboration
Composability
Governance
Natural Language Query
Data science integration and reporting are less relevant in this use case.
This use case focuses on organizations that aim to empower analytics content consumers, such as HR, sales, and operations teams, who directly consume analytical content to support business operations and decision-making.
Critical capabilities in this use case:
Natural Language Query
Data Storytelling
Automated Insights
Analytics Catalog
Data Visualization
Collaboration
Metrics Layer
Governance
Composability, data preparation, data science integration, and reporting are less relevant in this use case.
This use case mainly focus on enabling users to test hypotheses and build non-production models that can be handed over to data scientists or MLOps teams for deployment.
Critical capabilities in this use case:
Data Science Integration
Data Preparation
Metrics Layer
Automated Insights
Collaboration
Composability
Data VisualizationGovernance
Analytics catalog, data storytelling, and reporting are less relevant in this use case.
GenAI has profoundly impacted ABI product capabilities, user cases, and overall experiences in many ways. The following highlights key changes brought by GenAI:
Examples of capabilities:
Automatically match, link, analyze, tag, and annotate data to prepare it for transformation.
Identify sensitive attributes in datasets.
Automate repetitive transformations and integrations.
Provide recommendations to enhance data quality and richness.
Generate, debug, and convert code (Python, R, SQL, DAX) automatically, along with generating documentation.
Examples of use experiences:
Enable drag-and-drop operations, allowing users to combine data from various sources effortlessly.
Support the creation of analytical models, such as custom metrics, collections, groupings, and hierarchies.
Generate natural language descriptions for code and interfaces.
Create code for interacting with databases, scripts, or APIs using natural language commands.
Identify errors in code and perform seamless conversions between different programming languages.
Use generative AI (LLMs) to write interpretable code documentation.
Automate and accelerate DSML and AI processes, such as data profiling, quality checks, harmonization, modeling, manipulation, enrichment/inference, synthetic data generation, metadata development, and data cataloging.
Examples of capabilities:
Analyze key drivers to identify significant factors affecting outcomes.
Detect anomalies and outliers in datasets automatically.
Perform intelligent clustering and segmentation of data.
Conduct predictive analytics to forecast future trends and patterns.
Examples of user experiences:
Automatically generate insights for end users by identifying the most critical attributes in datasets.
Provide real-time notifications of anomalies or outliers based on user roles and business workflows.
Automatically discover clusters within datasets for better segmentation.
Use foundational methods like ARIMA to generate forecasts for continuous variables in datasets.
Display prediction errors to enhance interpretability and reliability of forecasts.
Examples of capabilities:
Automate the storytelling of data insights.
Generate narratives using natural language (narrative automation).
Examples of user experiences:
Create news-style data stories that combine headlines, narrative text, data visualizations, and audio/video content based on continuous monitoring.
Automatically generate and summarize written or spoken narratives to present a series of analytical insights.
All these functionalities can be enhanced by integrating large language models (LLMs) and offering natural language interaction for a more intuitive experience.
Examples of capabilities:
Enhanced data visualization for more insightful and interactive analysis.
Multi-experience user interfaces that adapt to various devices and use cases.
“What-if” scenario planning to explore potential outcomes.
Correlation and graph analysis for uncovering relationships within data.
Geospatial analysis for location-based insights.
Examples of user experiences:
Enable highly interactive dashboards and data exploration through direct manipulation of chart visuals.
Optimize user interfaces, interaction modes, and analytical features for multi-experience analytics, improving content consumption.
Deliver immersive analytical experiences with collaborative and 3D visualization interfaces to meet evolving use case demands.
Leverage data-driven insights through augmented reality (AR), mixed reality (MR), and virtual reality (VR) technologies for enhanced decision-making.
Examples of capabilities:
Question and answer functionality for intuitive data exploration.
Logical reasoning to provide deeper analytical insights.
Suggestions and auto-complete to assist users in forming queries or actions.
Synonym recognition and adaptive learning for improved query understanding.
Chatbots for interactive analytics.
Integration with large language models (LLMs) to enhance query handling and interpretation.
Examples of user experiences:
Natural Language Query (NLQ) enables business users to query data by typing or speaking business terms through a search interface or chatbot.
Some ABI vendors use keyword search, others employ natural language processing to translate terms into natural language questions, while some combine both approaches.
Certain use cases allow querying structured data, while others support semantic searches across multi-structured information.
Examples of capabilities:
Facilitate seamless communication among users.
Build a collaborative community ecosystem for shared insights.
Support multi-role environments to cater to diverse user needs.
Enable agile development for rapid iteration and adaptability.
Examples of user experiences:
Collaboration in ABI platforms involves fostering a cooperative ecosystem where users can annotate and share analytical content within a social media-like native experience. Collecting diverse perspectives on data is essential for building consensus and driving complex decision-making processes.
Examples of capabilities:
Guided model building to streamline the development process.
Generation, integration, and exploration of DSML functionalities.
Automated algorithm selection for optimal model performance.
Automated model tuning for improved accuracy and efficiency.
Automated model deployment and monitoring to streamline operationalization.
Explainable AI to enhance transparency and trust in model outputs.
Integration with R and Python for advanced data science workflows.
Examples of user experiences:
These capabilities empower both citizen and professional data scientists to enhance development and prototyping of composable DSML models. They enable deep integration with broader DSML toolchains, creating a seamless and efficient ecosystem for advanced analytics and machine learning.
Examples of capabilities:
Map metrics to business processes and organizational goals.
Publish, share, and push metrics to broader audiences to support actionable outcomes.
Use generative AI (GenAI) to connect ABI, DSML, and SaaS applications, providing data insights across platforms.
Examples of user experiences:
This capability introduces a virtualization layer that allows users to:
Define business metrics as code.
Manage these metrics directly from data warehouses.
Support downstream analytics, data science, and business applications.
With GenAI, more platforms are built on semantic layers, delivering data insights through natural language interactions for a seamless and intuitive user experience.
Examples of capabilities:
Embedded analytics for seamless integration into business workflows.
API and SDK support for customization and extensibility.
Operational frameworks to automate and streamline decision-driven workflows.
Decision-centric user interfaces to model, catalog, and audit decisions using metadata.
Codified analytics leveraging domain-specific languages (DSL) for workflow and logic representation.
Examples of user experiences:
This functionality focuses on assembling flexible, modular, and user-friendly ABI features by embedding enhanced analytics, utilizing APIs/SDKs, and implementing containerized or microservices architectures:
Embedded Analytics: Contextualize prescriptive analytics within business scenarios for actionable insights.
Operational Frameworks: Enable users to construct and automate data-driven decision workflows, triggering real-time business operations.
Decision-Centric User Interfaces: Facilitate modeling, cataloging, and auditing decisions using decision metadata.
Codified Analytics: Represent analytical workflows and logic as code or configuration files using DSL, allowing users to manage them with agile practices like any other software code.
GenAI is rapidly becoming a key accelerator for augmented analytics. It empowers users with limited technical expertise to ask highly complex business questions, thereby driving increased adoption of analytics. Gartner highlights the trend of "consumers becoming creators," reflecting the evolving role of business users in modern data and analytics (D&A)—shifting from passive insight consumers to active insight creators. This shift is driving rapid evolution and iteration in product design.
AI-powered IDE products like Cursor, Windsurf, and Bolt.New are revolutionizing the global software development landscape. However, the transformation of data analytics by AI is inherently more challenging due to the zero-tolerance for errors in the final stages of business decision-making. Despite this complexity, the trend is unfolding, and once breakthroughs are achieved, the disruptive impact on industries worldwide is expected to far surpass that of software development.
This article is a summary of insights gathered from numerous Gartner reports. Deep appreciation goes to Gartner analysts for their amazing works.