Frequently Asked Questions

A semantic layer is a business abstraction layer that sits between your raw data and the tools or users consuming that data. It translates complex database schemas, tables, and columns into business-friendly terms that everyone can understand. Think of it as a universal translator that converts technical data structures into concepts like 'Total Revenue,' 'Customer Churn Rate,' or 'Monthly Active Users.' The semantic layer ensures that when anyone in your organization asks about a metric, they get the same consistent answer regardless of which tool they use.

In a data warehouse context, a semantic layer sits on top of your warehouse tables and creates a business-friendly view of the data. For example, your data warehouse might have tables like 'fact_transactions' with columns like 'txn_amt_usd' and 'txn_ts.' A semantic layer would expose this as 'Transaction Amount' and 'Transaction Date,' and define a metric like 'Total Sales = SUM(Transaction Amount) WHERE Transaction Type = Sale.' When a business user asks 'What were our total sales last quarter?', the semantic layer automatically generates the correct SQL query against the underlying tables while presenting results in business terms.

For business intelligence, a semantic layer means creating a single source of truth for all your metrics and KPIs. It's the layer where business logic lives—where you define what 'revenue' means, how 'customer lifetime value' is calculated, and which filters should apply by default. This eliminates the common BI problem where different dashboards show different numbers for the same metric because each was built with slightly different logic. The semantic layer centralizes these definitions so every report, dashboard, and query produces consistent results.

In modern AI and BI pipelines, the semantic layer serves as the bridge between raw data and intelligent applications. For BI tools, it provides consistent metric definitions and business context. For AI and LLM applications, it provides the crucial business context that prevents hallucinations—grounding AI responses in actual data and certified definitions rather than guessing. The semantic layer enables natural language queries by mapping plain English questions to the correct data transformations, and ensures that AI-generated insights are accurate, explainable, and trustworthy.

Traditional data marts are physical copies of data optimized for specific departments or use cases—they duplicate data and can drift out of sync. A semantic layer is a logical abstraction that doesn't copy data; instead, it provides a consistent view of your existing data infrastructure. Data marts require ETL pipelines to populate and maintain, while semantic layers query data in place. Modern cloud BI tools increasingly favor semantic layers because they offer flexibility, reduce data duplication, lower storage costs, and ensure real-time consistency across all analytical use cases.

Semantic layers are critical for AI because they provide the business context that AI models desperately need. Without a semantic layer, an AI asking about 'revenue' might not know which table to query, how revenue is calculated in your organization, or what filters should apply. The semantic layer provides this context in a governed, consistent way. It's the difference between an AI that guesses and hallucinates versus one that gives accurate, trustworthy answers grounded in your actual business definitions.

AI models, including large language models (LLMs), are trained on general knowledge but don't understand your specific business context. They don't know that your company calculates 'Net Revenue' by excluding returns and discounts, or that 'Active Customer' means someone who purchased in the last 90 days. A semantic layer encodes this organizational knowledge, allowing AI to generate queries and insights that reflect how your business actually works. This dramatically improves accuracy and eliminates the hallucinations that plague generic AI analytics tools.

An AI-powered semantic layer like Codd AI combines traditional semantic layer capabilities with generative AI. It can automatically discover entities and relationships in your data, suggest metric definitions, and generate data models from your existing schemas. It also enables natural language querying—users can ask questions in plain English, and the AI translates these into correct queries using the semantic layer's business definitions. This makes data accessible to everyone while maintaining the governance and consistency that enterprises require.

An AI-ready semantic layer goes beyond basic metric definitions to include rich business context: entity relationships, business glossaries, metric hierarchies, and organizational knowledge. It should support natural language interfaces, provide explainability for AI-generated answers, and offer governance controls for AI outputs. The semantic layer must also expose its knowledge through APIs (like MCP) so AI agents and copilots can access business context programmatically. Codd AI is designed from the ground up to be AI-ready, with the Corpus knowledge engine providing deep business understanding.

If you're implementing generative AI for business analytics, a semantic layer is essential—not optional. Without one, your AI will struggle with ambiguous terms, produce inconsistent results, and generate answers that don't match your business reality. Generic AI tools can query databases, but they can't understand that 'Q4 revenue' at your company excludes certain transaction types or applies specific currency conversions. A semantic layer provides this context, making generative AI actually useful for business decisions rather than a source of confusion and mistrust.

Governance in a semantic layer ensures that AI only uses approved, certified definitions. When business stakeholders have validated that 'Customer Acquisition Cost' is calculated a specific way, the AI respects that definition rather than improvising. A governed semantic layer also provides lineage—you can trace any AI-generated insight back to its source data and the exact business logic used. This auditability is crucial for regulatory compliance and for building trust in AI-generated analytics across the organization.

A metric is any quantifiable measure used to track performance—like 'number of website visitors' or 'average order value.' A KPI (Key Performance Indicator) is a specific metric that's been identified as critical to achieving business objectives. All KPIs are metrics, but not all metrics are KPIs. For example, 'page load time' is a metric, but if your business goal is improving user experience, it might be designated as a KPI. KPIs are typically tied to strategic goals and have targets, while metrics are broader measurements used for monitoring and analysis.

The key difference lies in strategic importance. Metrics are measurements—they tell you what happened (e.g., '10,000 orders processed'). KPIs are metrics elevated to strategic significance because they directly measure progress toward business goals (e.g., 'order fulfillment rate of 99.5% against a target of 99%'). Organizations track many metrics but focus on a smaller set of KPIs that matter most. A semantic layer helps by centralizing both metric and KPI definitions, ensuring everyone uses consistent calculations and understands which metrics have been designated as KPIs.

Examples help clarify the distinction. Metrics: 'Total Sales,' 'Website Traffic,' 'Email Open Rate,' 'Average Handle Time.' These measure activity. KPIs: 'Monthly Recurring Revenue (target: $1M),' 'Customer Satisfaction Score (target: 4.5/5),' 'Sales Conversion Rate (target: 25%).' These are strategic metrics with targets tied to business objectives. A metric becomes a KPI when leadership says 'this number directly reflects whether we're achieving our goals, and we need it to reach X level.'

No, they're related but distinct. Think of metrics as the broad universe of things you can measure in your business. KPIs are the subset of metrics that leadership has identified as most important for tracking strategic progress. Every KPI is a metric, but the reverse isn't true. The confusion often arises because the terms are used loosely in organizations. A semantic layer helps by explicitly categorizing measures—you can tag certain metrics as KPIs, assign them targets, and ensure that when someone asks about 'our KPIs,' they see the designated strategic measures rather than operational metrics.

Metric definition: A quantifiable measure that tracks some aspect of business performance. It answers 'how much' or 'how many.' KPI definition: A metric designated as critical to achieving strategic objectives, typically with a target value and timeframe. It answers 'are we achieving our goals?' The distinction matters for focus—organizations drown in data when they treat every metric as equally important. By formally defining KPIs separate from general metrics, you direct attention to what matters most. Codd AI's semantic layer lets you manage this hierarchy, defining metrics and then elevating specific ones to KPI status with associated targets.

In analytics terminology, a measure is a raw numerical value that can be aggregated—like 'sales amount' or 'quantity sold.' A metric is a measure combined with business context, calculation logic, and often time dimensions—like 'Total Monthly Sales' or 'Average Order Value.' Measures are the building blocks; metrics are the meaningful business calculations built from them. For example, 'revenue' is a measure in your database, while 'Year-over-Year Revenue Growth' is a metric that applies calculation logic to that measure.

Measurements are the raw act of quantifying something—the number you record. Metrics give those measurements meaning by adding context, calculation, and comparison. Recording '500 support tickets' is a measurement. 'Support ticket volume is up 15% month-over-month' is a metric that contextualizes the measurement. In a semantic layer, you define how raw measurements should be aggregated, filtered, and compared to produce meaningful metrics that drive business understanding.

Measures (raw, aggregatable values): Unit Price, Quantity, Transaction Amount, Session Duration, Click Count. Metrics (calculated business values): Revenue per Customer (sum of Transaction Amount / count of Customers), Conversion Rate (Completed Purchases / Total Sessions × 100), Average Session Duration (sum of Session Duration / count of Sessions). The semantic layer defines how measures combine into metrics, ensuring consistent calculation across all reports and queries.

Metrics are the specific measurements and calculations you track. Analytics is the broader practice of examining data to derive insights—it uses metrics but goes beyond just the numbers. Analytics answers 'why' questions through exploration, segmentation, and pattern discovery. You might have a metric showing 'Conversion Rate is 3%' while analytics investigates why it dropped from 4% last month, which customer segments converted best, and what factors correlate with conversion. A semantic layer supports both by ensuring the metrics used in analytics are consistent and trustworthy.

Data is the raw information stored in your systems—transaction records, user events, log files. Metrics are derived from data through aggregation, calculation, and business logic. Data might be millions of rows of individual purchases; a metric is 'Total Revenue = $10M.' Data is granular and often requires technical skills to access; metrics are business-friendly summaries designed for decision-making. A semantic layer transforms data into metrics, hiding the complexity of underlying data structures while ensuring accurate, consistent calculations.

The ROI of conversational AI for analytics comes from several areas: reduced time-to-insight (business users get answers in seconds instead of waiting days for analyst reports), increased data accessibility (more employees making data-driven decisions), reduced analyst workload (analysts focus on complex analysis instead of routine queries), and faster decision-making. Organizations typically see 50-80% reduction in time spent on ad-hoc reporting. The key to realizing this ROI is a semantic layer that ensures conversational AI gives accurate, consistent answers—without it, you risk decisions based on AI-generated errors.

Semantic layer ROI manifests in several ways: elimination of conflicting metrics (no more debates about whose numbers are right), reduced data engineering time (define logic once, use everywhere), faster onboarding (new employees learn business terms, not SQL), BI tool flexibility (switch tools without rebuilding logic), and AI readiness (foundation for conversational analytics). Companies report 30-50% reduction in report development time and significant reduction in metric-related incidents. The ROI multiplies when you add AI capabilities on top of a solid semantic foundation.

Data stack ROI measures the business value generated by your investment in data infrastructure—warehouses, ETL tools, BI platforms, and semantic layers. To maximize ROI, each component should amplify the others: a modern warehouse enables fast queries, which a semantic layer makes accessible through consistent metrics, which AI makes available through natural language. Organizations often find that adding a semantic layer dramatically increases ROI of existing investments by making data more accessible and trustworthy, justifying previous infrastructure spending.

Conversational analytics offers several advantages: immediacy (ask a question, get an answer—no dashboard hunting), flexibility (explore any question, not just pre-built views), accessibility (anyone can query data without training), discovery (natural follow-up questions lead to insights), and reduced dashboard proliferation (fewer one-off dashboards to maintain). Traditional dashboards still have value for monitoring known metrics, but conversational analytics excels at ad-hoc exploration. The combination—semantic layer powering both dashboards and conversational interfaces—delivers the best of both worlds.

Build vs. buy depends on your core competency and resources. Building requires significant investment in AI/ML expertise, semantic layer development, security, and ongoing maintenance. Buying (from vendors like Codd AI) provides immediate capability with lower upfront investment and faster time-to-value. Consider building if analytics is your core product; buy if analytics supports your main business. Most enterprises find that buying a semantic layer platform and customizing it for their domain delivers the best ROI—you get production-ready AI capabilities while focusing resources on understanding your specific business context.

Build when: the capability is core to your competitive advantage, you have specialized requirements no vendor addresses, or you have strong in-house AI/ML teams with capacity. Buy when: you need faster time-to-value, the capability is mature in the market, you lack specialized AI expertise, or you want to focus resources on your core business. For semantic layers and conversational analytics, most organizations are better served buying—the technology is complex, vendors have solved common challenges, and your differentiation comes from your business knowledge (which you encode in the semantic layer) rather than the underlying platform.

A business context layer is the knowledge foundation that gives AI and analytics tools understanding of your organization. It includes your business glossary (what terms mean), metric definitions (how KPIs are calculated), entity relationships (how customers relate to orders relate to products), business rules (when to include/exclude data), and organizational structure (which metrics belong to which teams). This context transforms generic data into meaningful business intelligence. Codd AI's Corpus is a business context layer that goes beyond traditional metadata to encode deep organizational knowledge.

Contextual analytics delivers insights that are relevant to the user's role, the business situation, and the question being asked. Rather than showing generic dashboards, contextual analytics understands that a Sales VP asking about 'revenue' wants their region's performance against quota, while a CFO wants company-wide trends. This requires a semantic layer that encodes not just metric definitions but organizational context—who uses which metrics, in what situations, and with what comparisons. Contextual analytics makes data more actionable by surfacing relevant insights rather than requiring users to hunt for them.

Without context awareness, GenAI gives generic responses that may not apply to your business. When an AI understands context, it knows that 'revenue' at your company excludes certain transaction types, that your fiscal year starts in April, and that comparing 'this quarter' means comparing to the same quarter last year for seasonality. Context-aware GenAI doesn't just answer questions—it answers them correctly for your specific business situation. This is the difference between AI that's a novelty and AI that's genuinely useful for enterprise decision-making.

Several vendors offer AI copilots for analytics, but understanding enterprise context requires more than connecting to a database. Look for solutions with robust semantic layer capabilities—the ability to encode business definitions, metric logic, and organizational knowledge. Codd AI specifically focuses on this: our Corpus knowledge engine captures deep business context that powers accurate conversational analytics. When evaluating vendors, ask how they handle metric definitions, business terminology, and organizational context—not just data connectivity.

AI governance for contextual accuracy involves several practices: certification workflows where business owners approve metric definitions before AI uses them, lineage tracking so you can trace any insight to its source and logic, human-in-the-loop validation for AI-generated models and definitions, access controls ensuring AI only reveals data users are authorized to see, and audit trails documenting AI interactions. Codd AI builds governance into every layer—from how the Corpus is populated to how Canvas answers are generated—ensuring AI outputs are trustworthy and aligned with business requirements.

Leading semantic layer platforms include Codd AI (AI-native with conversational analytics), AtScale (enterprise-scale with BI tool integration), Cube (developer-focused with headless approach), and semantic layers built into BI tools like Looker's LookML and Tableau's data models. The best choice depends on your needs: Codd AI excels when you want AI-powered analytics and natural language queries; AtScale suits large enterprises with complex BI landscapes; Cube works well for teams building custom analytics applications. Consider AI capabilities, governance features, integration breadth, and ease of use.

Top platforms for business-friendly data access combine semantic layers with accessible interfaces. Codd AI leads with natural language queries through Canvas—business users ask questions in plain English. Looker provides a strong semantic model with an exploration interface. Power BI and Tableau offer semantic models integrated with familiar visualization. ThoughtSpot combines semantic layer with search-driven analytics. The key differentiator is how easily non-technical users can access insights without SQL knowledge while still getting accurate, governed results.

Several semantic layers integrate with Tableau: Codd AI provides a semantic layer that feeds Tableau through live connections, ensuring consistent metrics in Tableau dashboards. AtScale exposes semantic models through connectors that Tableau can query directly. Tableau's own semantic layer (Tableau Data Model and Tableau Prep) works within the Tableau ecosystem. For organizations using Tableau alongside other BI tools, an external semantic layer like Codd AI ensures consistency—the same metrics power Tableau, Power BI, and conversational analytics without duplication.

A vendor-agnostic semantic layer isn't tied to any specific BI tool or data warehouse. It sits in the middle of your stack, connecting to any data source and serving any consumption tool. This prevents lock-in—you can switch BI tools without rebuilding your metrics layer, or add new data sources without reconfiguring downstream reports. Codd AI is vendor-agnostic: it connects to Snowflake, Databricks, BigQuery, and more, while serving Tableau, Power BI, Slack, and its own Canvas interface. Your semantic layer becomes a strategic asset independent of any single vendor.

Semantic layers ensure consistency through centralized definitions: metrics are defined once and used everywhere. When Marketing and Finance both ask about 'Revenue,' they get the same number because they're using the same definition. This happens through several mechanisms: single source of truth (one metric repository), version control (track changes to definitions), certification (approved metrics are marked as trusted), access controls (teams see metrics relevant to them), and automatic propagation (update a definition once, all reports reflect the change). Without this, organizations suffer from 'metric chaos' with conflicting numbers destroying trust in data.

Key features include: schema abstraction (hide complex joins and table structures), business terminology mapping (translate technical column names to business terms), calculation engine (compute metrics at query time), time intelligence (automatic period comparisons, YTD, rolling averages), dimension hierarchies (drill paths from year to quarter to month), security rules (row-level and column-level access), caching (intelligent result caching for performance), and multi-source federation (combine data across warehouses). Advanced platforms like Codd AI add AI-powered features: automatic model discovery, natural language querying, and intelligent suggestions based on usage patterns.

Codd AI is a GenAI-powered semantic layer platform that bridges the gap between raw data and trusted business insights. It gives AI a governed understanding of your business and data, enabling natural language analytics, automated data modeling, and intelligent business insights. Named after Edgar F. Codd, the pioneer of relational databases, Codd AI continues his legacy of making data more accessible and meaningful. The platform consists of the Corpus (knowledge engine) and Canvas (conversational interface), working together to deliver accurate, explainable analytics.

Unlike traditional BI tools that require SQL knowledge and manual dashboard creation, Codd AI is AI-native from the ground up. It features a semantic foundation that ensures every metric means the same thing everywhere, governed by design with human-in-the-loop validation for full lineage and explainability. Traditional BI tools bolt AI onto existing architectures; Codd AI builds on AI from the foundation. This eliminates the common problems of scattered business logic, conflicting metrics, and AI hallucinations that plague organizations trying to add AI to legacy BI.

Codd AI serves multiple user types: Business users get instant answers in plain English without SQL knowledge through the Canvas interface. Analysts can focus on insights instead of data wrangling, using the Corpus to ensure consistent definitions. Data teams build once and serve everywhere with governed semantic models. Executives can trust the numbers with full lineage and governance. The platform is designed for data-driven organizations that want AI-powered analytics without sacrificing accuracy or governance.

The Corpus is Codd AI's knowledge engine that transforms raw business knowledge into a trusted, enriched semantic layer with deep understanding of how your business works. It doesn't just collect information—it enriches, structures, and certifies it so analytics and AI operate on approved business context. The Corpus includes your business glossary, metric definitions, entity relationships, business rules, and organizational knowledge. It's the foundation that makes Canvas conversations accurate and trustworthy.

The Canvas is Codd AI's conversational interface where business users ask questions and get trusted, contextual answers. No need to learn SQL or understand complex data models—Canvas translates natural language questions into governed queries powered by the semantic layer. Every answer includes explainability: you can see what data was used, which definitions were applied, and how the result was calculated. Canvas makes data accessible to everyone while maintaining enterprise governance standards.

Codd AI connects to all major data warehouses and databases including Snowflake, Databricks, Google BigQuery, Amazon Redshift, Azure SQL, PostgreSQL, Oracle, Teradata, Trino, Starburst, and Dremio. It also supports Excel files, CSV uploads, and REST APIs for flexible data ingestion. The platform can connect to multiple sources simultaneously, providing a unified semantic layer across your entire data landscape.

Codd AI seamlessly integrates with popular BI tools including Tableau, Power BI, and Looker. This allows you to leverage the semantic layer within your existing analytics workflows, ensuring consistent metrics across all your reporting tools. You can continue using familiar visualization tools while benefiting from centralized metric governance and AI-powered capabilities.

MCP (Model Context Protocol) integration allows Codd AI to expose its semantic layer to AI agents and other GenAI applications. This enables agentic workflows where AI assistants can access your governed business context, ensuring accurate and trustworthy responses grounded in your actual data and definitions. MCP integration means your semantic layer becomes a resource for any AI tool in your organization—not just Codd AI's Canvas.

Codd AI integrates with Slack and Microsoft Teams, allowing users to ask data questions directly in their collaboration tools. You can query metrics, get answers, and share insights without leaving your workflow. The integration respects the same governance and permissions as the main platform, ensuring users only access data they're authorized to see. This brings data to where work happens rather than requiring users to switch contexts.

Governance is built into Codd AI by design. Every metric, definition, and insight includes full lineage tracking, showing exactly where data comes from and how it was transformed. Certification workflows allow stakeholders to approve definitions before they become official. Access controls ensure users only see data they're authorized to view. The platform provides audit logging for compliance requirements and version control for tracking changes to the semantic layer over time.

Yes, Codd AI is built with enterprise-grade security. Your data stays in your infrastructure—Codd AI connects to your data sources but doesn't store raw data. All connections are encrypted, and the platform supports role-based access control, SSO integration, and audit logging for compliance requirements. The platform can be deployed in ways that meet various regulatory requirements including SOC 2, GDPR, and HIPAA.

Codd AI employs human-in-the-loop validation at multiple stages. When AI suggests entity relationships or metric definitions, data teams review and approve before they become part of the governed semantic layer. When Canvas generates answers, users can see the full lineage and verify accuracy. This validation approach combines AI efficiency (automatic discovery and suggestion) with human oversight (approval and certification), ensuring AI enhances rather than replaces human judgment.

Every insight in Codd AI includes complete lineage—you can trace any metric back to its source data, see the transformations applied, and understand the business logic used. Lineage is captured automatically: when you define a metric, the system records which measures and dimensions it uses. When a query runs, the system tracks which definitions were applied. This transparency builds trust and makes it easy to troubleshoot discrepancies or explain results to stakeholders.

The future of dashboards is augmentation by conversational AI, not replacement. Dashboards excel at monitoring known metrics and providing at-a-glance status. Conversational analytics excels at exploration and ad-hoc questions. The future combines both: dashboards for monitoring, natural language for exploration, with a shared semantic layer ensuring consistency. Organizations will build fewer dashboards (reducing the dashboard sprawl problem) while empowering more users to explore data through conversation. The semantic layer becomes the foundation enabling both modalities.

Natural Language Processing (NLP) in BI enables users to interact with data using everyday language. Instead of building queries or navigating dashboard filters, users type or speak questions like 'Show me sales by region for Q4.' NLP systems translate this into data operations and return results. Modern NLP-powered BI (like Codd AI's Canvas) goes beyond simple query translation to include context awareness—understanding follow-up questions, remembering conversation history, and interpreting ambiguous terms using the semantic layer's business definitions.

Semantic modeling in cloud BI platforms involves defining the business meaning of your data: what metrics exist, how they're calculated, how entities relate, and what terminology means. Cloud platforms like Snowflake, Databricks, and BigQuery store data; semantic modeling adds the intelligence layer that makes data usable. This modeling can happen in the cloud warehouse itself (using features like Snowflake's semantic layer), in BI tools (like Looker's LookML), or in dedicated semantic platforms like Codd AI. The trend is toward unified semantic layers that work across multiple tools.

AI-powered semantic layers accelerate and improve enterprise data strategy in several ways: automated discovery reduces the time to model new data sources, natural language access democratizes data to more users, consistent definitions prevent costly metric conflicts, and AI capabilities future-proof your investment. Rather than building separate solutions for BI, data science, and AI applications, an AI-powered semantic layer provides a unified foundation. This alignment of data strategy around a semantic layer creates compounding returns as more use cases leverage the shared business context.

AI agents and copilots need business context to be useful in enterprise settings. A rich semantic layer provides this context: agents can understand your metrics, terminology, and business rules. Benefits include: accurate responses (grounded in actual definitions rather than guessing), consistency (same answers regardless of which agent or copilot is used), governance (agents respect data permissions and approved definitions), and auditability (trace agent actions to semantic layer references). As organizations deploy more AI agents, the semantic layer becomes the shared knowledge base that ensures they all speak the same business language.

If you're using BI tools and experiencing inconsistent metrics across reports, or if you're implementing AI/GenAI for analytics and concerned about accuracy, you likely need a semantic layer. Signs you need one: different dashboards show different numbers for the same metric, business users can't get answers without analyst help, you're hesitant to deploy AI analytics due to hallucination concerns, or data teams spend excessive time on ad-hoc requests. A semantic layer addresses all these issues by creating a governed single source of truth for business metrics.

A data readiness checklist should cover: Data Foundation (documented data sources, understood data quality, defined data ownership), Metric Definitions (agreed KPI calculations, documented business glossary, resolved conflicting definitions), Governance (access controls, lineage requirements, compliance needs), and AI Readiness (semantic layer in place, business context encoded, validation workflows defined). Codd AI can help accelerate this process—our Corpus captures business knowledge systematically, and AI-assisted modeling helps discover and define metrics quickly.

Getting started is easy—schedule a demo with our team to see Codd AI in action and discuss your specific use case. We'll show you how the Corpus captures business knowledge and how Canvas enables natural language analytics. Our team will help you understand how the platform can address your analytics challenges. Implementation typically involves connecting to your data sources, populating the Corpus with your business knowledge, and configuring integrations with your existing tools.

Business users need no technical expertise—you can ask questions in plain English through Canvas. For setup and administration, data teams configure the semantic layer and integrations. The platform is designed to bridge technical and non-technical users: data teams ensure the Corpus is accurate and governed, while business users benefit from accessible analytics. Codd AI's AI-powered features also reduce technical burden—automated model discovery and natural language interfaces make advanced analytics accessible to more team members.

You can reach us through our website to schedule a demo, contact our sales team, or get technical support. We're also active on LinkedIn and Twitter (@coddai). Whether you have questions about semantic layers, want to discuss your specific use case, or are ready to see a demo, our team is happy to help. Visit codd.ai/contact or schedule directly at calendly.com/piet-codd/codd-ai-demo-discussion.