Magic Quadrant for Business Intelligence: Changes in 2025
Introduction: From Dashboards to Dialogue: How Generative AI and Composable Analytics are Revolutionizing Business Intelligence in 2025
During the past year, dramatic innovations in AI have fundamentally changed how users interact with their data, enabling chat-based interfaces and expanding data workflow capabilities. Additionally, platforms like Looker are moving beyond traditional BI with composable approaches that support data-driven workflows, alerts, applications, and more.
For more than 20 years, visual-exploration-based dashboards and reports have been the primary way users monitor and explore data in magic quadrant business intelligence solutions. However, the landscape is rapidly evolving as organizations seek more sophisticated ways to extract insights from their data assets. Today, modern analytics and business intelligence platforms are increasingly leveraging generative AI (GenAI) to enhance productivity for both developers and consumers of analytics content.
Organizations implementing these advanced ABI platforms typically realize several key benefits, including operational efficiency, enhanced scalability and flexibility, improved collaboration and communication, as well as managed governance and trust. The 2025 Magic Quadrant reflects these shifts, recognizing vendors who are successfully navigating the transition from static reporting to AI-augmented, interactive analytics environments.
Table of Contents
Redefining the ABI Market in 2025
The analytics and business intelligence (ABI) market is undergoing a fundamental transformation in 2025, moving beyond traditional reporting paradigms toward more intelligent, autonomous, and composable architectures. This evolution represents a significant shift in how organizations leverage data for strategic decision-making.
Shift from Dashboard-Centric to Agentic Analytics
The traditional dashboard-centric approach to business intelligence is rapidly giving way to agentic analytics, which fundamentally changes how organizations interact with their data. Unlike conventional BI systems that wait for users to open dashboards and examine reports, agentic analytics flips this model by enabling AI agents to autonomously monitor data streams, identify anomalies or opportunities, and decide how to respond without continuous human prompting [1].
These AI agents don’t merely support decisions—they execute and orchestrate actions, making them particularly valuable for organizations seeking to reduce decision latency [1]. True agentic analytics delivers the right insight to the right user at the right moment, enabling timely, meaningful action where users work [1]. In an agentic enterprise, agents understand goals, interpret changes, suggest responses, and learn from human guidance—they’re not just automating tasks [1].
Notably, agentic systems must be grounded in business logic and KPIs they’re meant to optimize [1]. Without this context, triggers alone can create noise rather than actionable intelligence. This represents the evolution from human-in-the-loop analytics to human-on-the-loop orchestration, where analysts guide strategy while machines handle operational intelligence at scale [1].
Role of Generative AI in ABI Platforms
Generative AI serves as a foundational platform enabling a wide variety of software applications and services in the ABI market [2]. It allows organizations to do more with their analytics platforms, such as generating data stories, metadata, code, executive summaries, and storyboards [2].
One of the most significant capabilities generative AI brings to ABI platforms is natural language interaction with data. Users can interrogate data using conversational language, receiving automated insights in real-time without traditional dashboards [2]. This advancement is made possible through integration with proprietary and third-party Large Language Models (LLMs) from providers like OpenAI, Microsoft, and Google [2].
Organizations implementing generative AI in their data processes report a 20-40% reduction in decision-making time, stemming from AI’s ability to analyze data continuously and deliver actionable insights promptly [3]. Moreover, AI-driven insights help reduce human biases in analysis, supporting more objective business decisions [3].
Increased Demand for Composability and Semantic Layers
Composable analytics has emerged as a crucial approach that enables organizations to combine modular components from different data and analytics tools instead of relying on monolithic systems [4]. This methodology allows businesses to break down data silos by integrating diverse data sources and technologies to create an analytics stack adaptable to changing requirements [4].
The benefits of composable analytics include:
- Increased flexibility and agility to adapt quickly to changing business needs
- Customization for specific business requirements rather than being constrained by off-the-shelf products
- Faster insights through combining diverse data sources and tools
- Wider access to BI data through low-code and no-code options [4]
Underpinning effective composable analytics is the semantic layer, which has become essential for trustworthy AI and analytics. The semantic layer sits between raw data and downstream tools, translating complex SQL into consistent, governed business terms [5]. It provides a consistent understanding and interpretation of data across the organization and its applications [4].
In fact, tests show that without a shared business context, LLMs can be wrong 80% of the time when answering business questions, but when grounded in a semantic layer, they achieve near-perfect accuracy [6]. Consequently, the semantic layer is no longer optional—it’s foundational for giving generative AI and every analytics tool access to governed, contextualized, and business-aligned data [6].
Changes in the Magic Quadrant Evaluation Criteria
Gartner’s Magic Quadrant for Analytics and Business Intelligence (ABI) platforms has evolved significantly in 2025, reflecting fundamental shifts in how vendors are evaluated. These changes highlight the growing importance of advanced capabilities that enable more intuitive, trustworthy, and integrated analytics experiences.
Expanded Focus on Natural Language Query (NLQ) and NLG
Natural language capabilities have become central evaluation criteria in the 2025 Magic Quadrant, marking a significant evolution from previous years. NLQ now allows users to interact with data using everyday language through text or voice interfaces, eliminating the need for technical expertise in formulating complex queries [7]. This democratization of analytics, referred to by Gartner as “democratizing analytics,” makes tools directly accessible to business users beyond specialized analytics professionals [7].
Advanced NLQ features now being evaluated include:
- AI-assisted query formulation that presents contextual query suggestions to improve question quality and minimize errors [8]
- Integration with large language models (LLMs) that enable more sophisticated understanding of user intent and context [8]
- Ability to maintain conversational context across multiple queries [9]
Furthermore, natural language generation (NLG) capabilities have gained prominence in evaluation criteria. NLG technology enables BI tools to automatically generate narratives from data, describing trends, variances, and exceptions in plain language [7]. This functionality significantly reduces the time business analysts spend creating data narrations and ensures consistent interpretation of visualizations [7].
Governance and Trust as Core Evaluation Pillars
Throughout 2025, data governance has transitioned from a supplementary consideration to a core evaluation pillar in the Magic Quadrant assessment. Ethical data governance frameworks now factor prominently in vendor evaluations, focusing on policies and structures that ensure data quality, security, and ethical considerations related to data collection and use [10].
Microsoft’s positioning as a Leader in the 2025 Magic Quadrant partly stems from its introduction of a centralized Fabric metrics layer within Power BI, enabling organizations to define, discover, and reuse trusted metrics across reports [11]. This approach fosters data consistency, reduces duplicated metric definitions, and strengthens governance by ensuring standardized calculations [11].
Subsequently, vendors demonstrating robust governance capabilities must now provide granular access controls, comprehensive audit trails, and built-in compliance frameworks [12]. This shift is especially relevant for organizations in regulated industries such as healthcare, finance, and government, where compliance requirements are stringent [12].
Integration with Cloud Ecosystems and Business Applications
Integration capabilities have emerged as decisive evaluation criteria in the 2025 Magic Quadrant. Specifically, vendors must demonstrate how effectively their platforms connect with modern data stack components and business applications [12]. This evaluation dimension recognizes that standalone BI tools no longer suffice—they must function seamlessly within broader enterprise ecosystems.
Google Looker’s robust open architecture and semantic layer contribute to its strong positioning, offering extensive integration capabilities through APIs that allow users to identify data sources effectively [1]. Simultaneously, Oracle Analytics Cloud stands out for its close integration with Oracle Fusion Data Intelligence, providing pre-packaged data analytics and models for horizontal applications spanning finance to CRM [1].
Essentially, organizations seeking optimal BI solutions must assess whether vendors offer robust APIs, white-label solutions, and embedding capabilities that align with their application ecosystem [12]. The ability to handle streaming data sources, provide low-latency analytics, and support operational decision-making with minimal delay between data generation and insight delivery has become crucial [12].
The 2025 Magic Quadrant accordingly places greater emphasis on cloud-native architectures, recognizing the shift toward cloud-based BI solutions that provide organizations with scalable, flexible, and real-time access to critical business data [10].
What Changed for the 2025 Market Leaders
The leading vendors in the magic quadrant business intelligence market have made substantial advancements in 2025, aligning with the industry’s shift toward AI-powered analytics and composable architectures.
Microsoft Power BI: Copilot Integration and Fabric Expansion
Microsoft has strengthened its position through the integration of Copilot and expansion of Microsoft Fabric. Copilot in Power BI represents a significant leap forward, allowing users to generate visualizations, create and edit DAX calculations, and produce narrative summaries using conversational language [13]. The May 2025 update introduced a standalone Copilot feature that enables users to “Ask Anything!” across reports, semantic models, and data agents they have access to [14]. Additionally, Microsoft has standardized on open data formats by adopting Delta Lake and Parquet as native storage formats, helping organizations avoid vendor lock-in [13]. The introduction of Direct Lake mode delivers import-mode performance with DirectQuery’s real-time capabilities [13].
Google Looker: Gemini and Semantic Layer Enhancements
Google has enhanced Looker with Gemini integration and semantic layer improvements. Gemini in Looker provides conversational analytics capabilities, allowing users to explore data by asking questions in natural language [15]. The semantic layer acts as a foundation for trustworthy gen AI, reducing data errors in natural language queries by as much as two-thirds [16]. Internal testing confirms that Looker’s semantic layer is crucial for accuracy, serving as a centralized data dictionary that defines business logic and relationships [17]. Google also released new BI connectors, including general availability of a custom-built connector for Tableau, plus an Open SQL Interface that gives customers more options for deploying governed analytics [18].
Salesforce Tableau: Launch of Tableau Next and Agentic AI
Salesforce has introduced Tableau Next, a flexible, API-first analytics experience that integrates with Agentforce, creating a new paradigm for business intelligence [2]. This platform introduces agentic analytics where humans collaborate with AI agents to transform data workflows [19]. Tableau Next comes with three pre-built analytics skills: Data Pro (data preparation assistant), Concierge (natural language analysis), and Inspector (proactive data monitoring) [2]. The platform also features Tableau Semantics, an AI-infused semantic layer deeply integrated with Data Cloud that powers trusted, unified business data and accelerates semantic model creation [3].
Oracle Analytics: Fusion Data Intelligence and AI Assistant
Oracle has evolved its analytics offerings with Fusion Data Intelligence platform and AI Assistant capabilities. The Fusion Data Intelligence platform brings together business data, ready-to-use analytics, and prebuilt AI/ML models to deliver deeper insights [20]. Unlike competitors offering empty platforms, Oracle provides a complete data intelligence platform with turnkey data, analytics, and AI/ML content by domain [20]. The Oracle Analytics AI Assistant helps users build and refine visualizations through natural language prompts, with recent updates enabling integration of third-party large language models such as OpenAI’s ChatGPT-4 Turbo [21].
Qlik: Associative Model and Real-Time Streaming via Upsolver
Qlik has strengthened its position through the acquisition of Upsolver, expanding its real-time streaming capabilities. This acquisition delivers low-latency ingestion and optimization for Apache Iceberg, enabling continuous ingestion from streaming sources like Kafka and Kinesis [22]. The integration allows organizations to capture data as it’s generated and make it instantly available for analytics or AI models [4]. One customer reports ingesting 2 million events per second, bringing in 1 petabyte of raw data per month with no data engineers [4]. Additionally, Qlik’s Adaptive Iceberg Optimizer reduces storage and improves query performance by up to 5x through automated, table-specific optimizations [22].
Technology Trends Shaping ABI Platforms
Several key technological advancements are reshaping analytics and business intelligence platforms in 2025, enabling more sophisticated, accessible, and efficient data workflows.
Metrics Layer as a Universal Semantic Framework
The universal semantic layer has emerged as a critical foundation for modern magic quadrant business intelligence solutions. Functioning as a translation layer between complex data structures and business language, it provides consistency in metric definitions across organizations. This intermediary layer enables data producers to define metrics once and establish them as canonical, thereby eliminating inconsistencies and errors that occur when business users define metrics independently [23]. Beyond raw calculations, semantic layers incorporate vital contextual information about how metrics were derived and by whom, preserving the historical trajectory of data for proper interpretation [23]. Primary benefits include reduced computational costs, as semantic layers eliminate duplicate data transformations that typically consume a majority of a company’s compute budget [23]. These frameworks likewise foster data democratization by providing equal access to information across organizational hierarchies [23].
Composable Analytics and Embedded Workflows
Composable analytics represents a fundamental shift in business intelligence architecture, enabling organizations to combine specialized tools that work together seamlessly. Unlike monolithic systems, this approach allows for piecing together best-in-class components that connect cleanly and evolve as needs change [24]. Fundamentally, it creates a collaborative workspace where tools connect clearly, and each element can be improved without disrupting the entire system [24]. Applications embedded with business intelligence capabilities enable users to analyze data without switching between multiple tools, creating more seamless analytics workflows [25]. This integration reduces the time spent moving between business applications and analytics tools, creating more time for value-added activities and improving user adoption rates [25]. Additionally, embedded analytics reduces workload on analytics teams by providing business users with self-service capabilities while allowing specialists to focus on developing new products [25].
NLQ and NLG Enhancements Across Platforms
Natural language capabilities have become central to modern analytics platforms. AI-powered Natural Language Query (NLQ) allows users to ask questions in plain language through text or voice interfaces, eliminating the need for technical expertise or complex query languages [6]. Recent innovations include AI NLQ, which automatically generates natural language queries into accurate syntax, helping users formulate complex business queries for deeper insights [6]. These systems present contextual query suggestions to improve question quality and minimize errors [5]. Correspondingly, Natural Language Generation (NLG) technology enables BI tools to automatically create narratives from data, describing trends and exceptions in plain language [26]. This capability reduces the time business analysts spend creating data narrations and ensures consistent interpretation of visualizations [26].
Serverless Architectures and In-Memory Engines
Serverless architecture for Business Intelligence represents a significant architectural evolution, leveraging cloud-based services to build, deploy, and manage BI solutions without dedicated infrastructure [27]. Unlike traditional systems requiring dedicated servers and manual scaling, serverless BI operates on an event-driven model where resources are allocated dynamically based on demand [27]. Key benefits include cost efficiency through pay-as-you-go pricing, rapid deployment of BI solutions, and automatic scaling to handle fluctuating workloads [27]. Furthermore, with infrastructure management handled by cloud providers, IT teams can redirect focus from maintenance to data analysis and strategy [27]. This approach proves particularly valuable for businesses with unpredictable traffic patterns, trigger-based tasks, and those building RESTful APIs [28].
Strategic Implications for Buyers and Vendors
Organizations adopting analytics and business intelligence platforms face crucial strategic decisions that impact long-term success and return on investment. These considerations extend beyond mere feature comparisons to fundamental business strategy questions.
Vendor Lock-In vs. Open Ecosystem Strategies
Selecting between open architecture and proprietary systems represents a critical decision point for organizations implementing magic quadrant business intelligence solutions. Vendor lock-in occurs when the cost or effort of migrating to a different platform exceeds the benefits, despite compelling business reasons to switch [11]. Organizations trapped in this state face limitations in scalability, technical innovation, and integration capabilities. Open architecture alternatives provide architectural flexibility that protects against making suboptimal decisions, as future adjustments can be implemented with minimal disruption [11].
Yet, many cloud vendors design their offerings as “walled gardens,” making extraction difficult through egress fees and tight infrastructure dependencies [12]. To mitigate these risks, organizations increasingly adopt multi-cloud strategies that leverage multiple providers to prevent dependency on a single vendor [29]. This approach enables businesses to select optimal features from each provider, potentially resulting in significant cost savings alongside improved performance and flexibility [29].
Cost Transparency and Licensing Complexity
For most organizations, expenses remain the primary factor influencing IT budget decisions [30]. Without transparent cost structures, technology expenses can escalate rapidly, particularly as systems require updates or replacements. IT cost transparency—tracking the total expenditure required to deliver and maintain services—helps organizations ensure business growth isn’t hindered by IT budget constraints [30].
Various licensing models further complicate the landscape. Enterprise organizations typically structure their software license management through different approaches: centralized purchasing and management, centralized purchasing with decentralized management, or fully decentralized models [31]. Despite differences, organizations implementing cost transparency generally experience 15-20% cost savings across labor, contracts, and vendors [32].
Talent Availability and Community Support
Availability of skilled professionals significantly influences platform selection decisions. According to client feedback, certain platforms like Tableau are frequently praised specifically for talent availability—”any data professional we hire is very likely to already know Tableau” [33]. This advantage reduces training costs and accelerates implementation timelines.
Open source alternatives offer active community engagement where users share code, applications, and skills [7]. These communities provide valuable support resources, although they lack the formal support structures of commercial offerings.
Vertical-Specific ABI Solutions and Use Cases
Industry-specific analytics solutions are gaining prominence, offering pre-packaged capabilities tailored to particular sectors. Unlike traditional machine vision applications requiring customization for specific products, newer vertical solutions are designed for entire industries [8]. These solutions enable operational analytics across retail, hospitality, parking management, and other sectors [8].
Vertical AI agents built for specific industry workflows represent the future direction, automating repetitive tasks through context-aware systems [9]. These solutions deliver highest value when targeting repetitive, high-volume tasks with structured data and clear ROI metrics [9].
Conclusion
The 2025 Magic Quadrant for Business Intelligence reflects a fundamental transformation in how organizations derive value from their data assets. Traditional dashboard-centric approaches have undoubtedly given way to more sophisticated, AI-driven analytics systems capable of autonomous operation and proactive insight generation. These agentic analytics platforms now monitor data streams independently, identify anomalies, and execute actions without continuous human intervention.
Generative AI has emerged as the cornerstone technology driving this evolution, enabling natural language interactions with data and reducing decision-making time by 20-40%. This shift toward conversational analytics has consequently reshaped evaluation criteria, with capabilities like Natural Language Query and Natural Language Generation becoming essential benchmarks rather than optional features.
Market leaders have adapted accordingly to these changing demands. Microsoft strengthened its position through Copilot integration and Fabric expansion, while Google enhanced Looker with Gemini capabilities and semantic layer improvements. Salesforce introduced Tableau Next with agentic AI functionality, Oracle developed its Fusion Data Intelligence platform, and Qlik expanded real-time streaming capabilities through its Upsolver acquisition.
The universal semantic layer now stands as the critical foundation for trustworthy AI and analytics, serving as the translation layer between complex data structures and business language. Without this shared business context, large language models often produce inaccurate results when answering business questions. Therefore, semantic layers have transitioned from optional components to essential infrastructure.
Organizations must carefully weigh strategic considerations when selecting analytics platforms. The balance between vendor lock-in and open ecosystem strategies presents significant long-term implications. Similarly, cost transparency, talent availability, and alignment with industry-specific requirements deserve thoughtful analysis before committing to any platform.
The future of business intelligence clearly belongs to composable, AI-enhanced systems that seamlessly integrate with business workflows. These platforms will continue evolving toward more autonomous operation while maintaining human oversight, ultimately transforming how organizations leverage data for competitive advantage. Success in this new paradigm depends not merely on adopting advanced technology but on strategically implementing it within a comprehensive data governance framework that ensures trust, accuracy, and accessibility.
The 2025 Magic Quadrant for Business Intelligence reveals a dramatic shift from traditional dashboards to AI-powered, autonomous analytics systems that fundamentally change how organizations interact with data.
• Agentic analytics replaces dashboard-centric BI – AI agents now autonomously monitor data, identify opportunities, and execute actions without continuous human prompting, reducing decision-making time by 20-40%.
• Semantic layers become essential infrastructure – Without shared business context, LLMs are wrong 80% of the time, but semantic layers enable near-perfect accuracy for AI-driven insights.
• Natural language capabilities drive vendor evaluation – NLQ and NLG are now core assessment criteria, democratizing analytics by allowing business users to interact with data through conversational interfaces.
• Market leaders embrace composable architectures – Microsoft’s Copilot integration, Google’s Gemini enhancements, and Salesforce’s Tableau Next demonstrate the shift toward modular, API-first analytics platforms.
• Governance and trust become competitive differentiators – Vendors must provide robust data governance frameworks, granular access controls, and comprehensive audit trails to succeed in regulated industries.
The transformation represents more than technological advancement—it’s a fundamental reimagining of how organizations leverage data for competitive advantage. Success requires strategic implementation within comprehensive governance frameworks that ensure trust, accuracy, and accessibility across the enterprise.
FAQs
Q1. What major changes are occurring in the business intelligence market in 2025? The BI market is shifting from dashboard-centric approaches to AI-driven, agentic analytics. Platforms now leverage generative AI for natural language interactions, autonomous monitoring, and proactive insights generation, reducing decision-making time by 20-40%.
Q2. How are semantic layers impacting business intelligence platforms? Semantic layers have become essential infrastructure for trustworthy AI and analytics. They provide a consistent business context for data interpretation, significantly improving the accuracy of AI-generated insights and ensuring data consistency across organizations.
Q3. What new capabilities are being prioritized in the 2025 Magic Quadrant evaluation? The 2025 Magic Quadrant emphasizes natural language capabilities (NLQ and NLG), robust data governance frameworks, and seamless integration with cloud ecosystems and business applications as key evaluation criteria for BI platforms.
Q4. How are market leaders adapting to the evolving BI landscape? Market leaders are embracing AI integration and composable architectures. For example, Microsoft has integrated Copilot into Power BI, Google has enhanced Looker with Gemini capabilities, and Salesforce has introduced Tableau Next with agentic AI functionality.
Q5. What strategic considerations should organizations keep in mind when selecting a BI platform? Organizations should carefully consider the balance between vendor lock-in and open ecosystem strategies, evaluate cost transparency and licensing models, assess talent availability and community support, and determine alignment with industry-specific requirements before committing to a BI platform.
References
[1] – https://www.cxtoday.com/customer-data-platform/gartner-magic-quadrant-for-analytics-and-business-intelligence-abi-platforms-2025-the-rundown/
[2] – https://www.salesforce.com/analytics/tableau-next/
[3] – https://www.tableau.com/products/tableau-next
[4] – https://www.qlik.com/blog/embracing-iceberg-with-scalable-open-data-solutions-for-customer-success
[5] – https://www.yellowfinbi.com/blog/what-is-natural-language-query-nlq
[6] – https://www.prnewswire.com/news-releases/yellowfin-launches-ai-powered-natural-language-query-ai-nlq-in-version-9-15–enhancing-self-service-business-intelligence-302450003.html
[7] – https://www.domo.com/learn/article/open-source-bi-tools
[8] – https://www.abiresearch.com/market-research/insight/7785210-vertical-specific-video-iot-solutions-driv
[9] – https://www.abilytics.com/the-age-of-vertical-ai-agents/
[10] – https://improvado.io/blog/business-intelligence-trends
[11] – https://www.yellowfinbi.com/blog/how-to-avoid-bi-vendor-lock-in-with-open-architecture
[12] – https://drj.com/industry_news/understanding-the-risks-of-cloud-vendor-lock-in/
[13] – https://powerbi.microsoft.com/en-us/blog/introducing-microsoft-fabric-and-copilot-in-microsoft-power-bi/
[14] – https://powerbi.microsoft.com/en-ca/blog/power-bi-may-2025-feature-summary/
[15] – https://cloud.google.com/gemini/docs/looker/overview
[16] – https://cloud.google.com/blog/products/business-intelligence/how-lookers-semantic-layer-enhances-gen-ai-trustworthiness
[17] – https://www.googlecloudcommunity.com/gc/Modeling/Unlock-Data-Insights-with-Looker-s-Semantic-Model/m-p/886158
[18] – https://cloud.google.com/blog/products/business-intelligence/opening-up-the-looker-semantic-layer
[19] – https://www.tableau.com/blog/agentic-analytics-new-paradigm-for-business-intelligence
[20] – https://www.oracle.com/business-analytics/fusion-data-intelligence-platform/
[21] – https://blogs.oracle.com/analytics/post/oac-mar-2025-update-blog
[22] – https://www.qlik.com/us/news/company/press-room/press-releases/qlik-acquires-upsolver-to-deliver-low-latency-ingestion-and-optimization-for-apache-iceberg
[23] – https://www.getdbt.com/blog/universal-semantic-layer
[24] – https://www.sigmacomputing.com/blog/composable-analytics
[25] – https://www.techtarget.com/searchbusinessanalytics/feature/Embedded-BI-and-analytics-apps-speed-workflows-insight-access
[26] – https://www.techtarget.com/searchbusinessanalytics/feature/4-ways-natural-language-querying-in-BI-tools-can-benefit-users
[27] – https://www.meegle.com/en_us/topics/serverless-architecture/serverless-architecture-for-business-intelligence
[28] – https://www.datadoghq.com/knowledge-center/serverless-architecture/
[29] – https://www.seagate.com/blog/how-to-avoid-vendor-lock-in/
[30] – https://www.bmc.com/blogs/it-cost-transparency/
[31] – https://www.revenera.com/blog/software-monetization/managing-software-licenses-and-entitlements-in-complex-organization-hierarchies/
[32] – https://www.apptio.com/solutions/itfm/cost-transparency/
[33] – https://www.tableau.com/learn/whitepapers/forrester-wave-business-intelligence-report
