May 2026 Summaries

Over the past year, Sigma's Data Platform Team has integrated Snowflake Semantic Views into their analytics processes to enhance their semantic layer, allowing for streamlined metric definitions and improved data model interoperability with Sigma. By using dbt to manage these definitions, the team can leverage version control, peer review, and CI checks, ensuring consistency and traceability across their data models. Two main integration patterns are employed to incorporate these semantic views into Sigma: native integration for direct use of Snowflake Semantic Views as Sigma data models, and a more flexible API path for extending data models with Sigma-specific functionalities. Collaboration with non-technical stakeholders is facilitated through Sigma's platform, enabling them to interact with data models, provide feedback, and drive the evolution of the semantic layer. This structured approach has reduced ad-hoc modeling work, centralized metric definitions, and allowed various departments to access a consistent understanding of their data, ultimately promoting a cohesive and scalable data strategy.

Sigma is introducing Sigma Tables, a new feature that enhances data writeback capabilities in cloud data warehouses like Snowflake, Databricks, and Postgres, enabling shared access across multiple applications and workbooks. Unlike the previous Input Tables, which were limited to individual workbook usage, Sigma Tables allow complex workflows by facilitating read and write operations by various applications and systems to a single table, thereby creating a unified source of truth. This innovation allows business leaders, customer success managers, and operations teams to interact with the same data seamlessly, enhances integration with external systems, and supports the migration of SaaS and proprietary applications into Sigma, reducing costs and complexity. Sigma Tables maintain the established governance and security protocols of the underlying warehouse, ensuring data access policies are enforced at query time, while providing comprehensive auditability and centralized control for administrators. As Sigma Tables roll out for Snowflake, Databricks, and Postgres, users can join the early access program to explore the potential of this feature to consolidate workflows and enhance application connectivity within their data infrastructure.

Sigma Public is a newly launched, free platform that enables users to build, share, and explore data assets without requiring a company login or paid license, democratizing access to Sigma’s AI Apps and analytics tools. Previously, such capabilities were limited to those with a Sigma license, but now anyone can leverage Sigma Public to experiment with building AI apps, agentic analytics, and data visualizations using either provided sample data or their own uploads. This platform supports a hands-on approach to learning, aiding personal interest-driven skill development, and offers a unique opportunity for individuals to showcase their technical capabilities to potential employers. Sigma Public also fosters community engagement through challenges like Workout Wednesday, where users can recreate and learn from others' projects. This initiative is designed to inspire creativity and collaboration within the data community, encouraging experimentation and knowledge sharing among users, and providing a space for building portfolios or settling debates with data-driven insights.

B2B SaaS products often lose value after onboarding because they don't provide users a platform to build and explore analytics, driving customers to external tools like spreadsheets for their analytical work. The value of embedded analytics lies in its ability to create a sticky product experience by allowing customers to build and save custom views, reports, and workflows within the product, which in turn raises the switching costs for customers. Products with embedded analytics that go beyond static dashboards enable customers to interact with and act on their data, fostering a deeper, more integrated user relationship and making it difficult for them to switch to competitors. Companies like Mindbody, Astronomer, and HyperFinity have demonstrated that enabling customers to build within the product can lead to increased engagement, sales growth, and a stronger competitive position. The decision for SaaS teams to build or buy an analytics layer involves weighing the initial build costs against ongoing maintenance and adaptability to user needs, as internal builds often struggle to keep pace with evolving requirements, whereas embedding a mature platform offers immediate and compounding value.

Embedding analytics into products benefits both customers and companies, as it provides insights into customer engagement and helps identify opportunities for retention and expansion. While many B2B SaaS companies use analytics layers to serve customers, they often overlook the data's potential to enhance retention workflows by not connecting these insights to customer success strategies. By analyzing patterns such as declining dashboard activity, high CSV exports, high query volume, and permission boundary hits, companies can detect churn risks and expansion readiness. Sigma's native usage dashboard and audit logging features allow organizations to monitor and analyze these behaviors, turning raw usage data into actionable insights. This enables customer success and product teams to preemptively address issues or opportunities, thereby increasing customer satisfaction and operational efficiency. Companies like Astronomer and Mindbody have successfully leveraged these insights to improve customer engagement and drive growth, demonstrating the value of systematically integrating analytics signals into business strategies.

In recent years, Sigma has emerged as a key platform facilitating the rise of "vibe coding," a trend where employees across different departments create software applications without extensive IT oversight, using AI tools to quickly develop and deploy solutions tailored to their needs. This shift represents a significant change in how enterprise software is developed, with Sigma providing a secure and governed environment that maintains company-wide data governance while enabling rapid development. The platform's tools, such as Sigma Agents and Sigma Assistant, empower users to create data-driven applications and workflows directly from natural language prompts, leveraging the deep business context embedded in Sigma to produce more accurate outputs. As companies navigate this new landscape, Sigma is focused on enhancing its runtime environment to make it more intuitive and efficient, thus encouraging innovation within a secure framework. By automating model selection and integrating coding agents, Sigma aims to streamline the development process, allowing employees to build solutions with greater autonomy while ensuring compliance and auditability.

Sigma Systems has launched the Sigma SI Partner Program to structure and recognize the capabilities of nearly 200 global partners involved in deploying Sigma, building AI Apps, and developing Sigma Agents. This program, initiated in February 2026, categorizes partners into three tiers based on market reach, technical certifications, and solutions built on Sigma, with Elite and Premier partners demonstrating the highest levels of expertise and performance. The first cohort of Elite and Premier partners includes companies such as Aimpoint Digital, InterWorks, phData, and others, who have shown exceptional capability in leveraging Sigma for analytics projects. These partners provide strategic advisory, technical implementation, and enablement to help clients maximize the impact of their AI analytics investments. The program aims to enhance Sigma's ecosystem by fostering growth through cloud data warehouse-native analytics projects and facilitating AI-powered workflow development. Future enhancements to the program will include specializations to recognize partner capabilities in specific domains, such as AI Apps and embedded analytics.

Sigma Assistant is a new AI interface launched by Sigma, designed to enable businesses to analyze data and build applications using natural language while maintaining data governance and security. It integrates the functionalities of Ask Sigma and AI Builder, allowing teams to work with real business data within existing workflows without the risks associated with external data tools. Sigma Assistant facilitates data exploration by providing insights based on organizational data models, metrics, and usage patterns. It allows users to ask questions, follow up on insights, and iteratively build dashboards or applications, ensuring every AI-generated answer is verifiable and traceable back to the data source. This tool operates within a governed workspace, respecting warehouse permissions and security controls, and supports integration with various AI providers. Sigma Assistant leverages existing work and data models to enhance productivity and aims to expand its capabilities by incorporating richer business context and more complex app elements in future updates.

Embedded analytics projects often struggle to scale due to foundational decisions made before implementation, such as data model design, tenant isolation, performance architecture, and licensing. A successful data model must be intuitive for users, perform data transformations before data reaches the analytics layer, and incorporate tenant isolation from the outset to support multi-tenancy and self-service exploration. Tenant isolation can be achieved through row-level security or connection-level isolation, each with its trade-offs regarding security and operational overhead. Performance issues usually arise from decisions about data model shape, warehouse partitioning, clustering strategy, and query concurrency configuration, which are costly to reverse once in place. Materializing results is beneficial for computationally expensive queries with stable underlying data, while real-time querying is best for scenarios requiring up-to-date information. Planning these elements early, alongside a thorough understanding of licensing costs, can prevent expensive rework and ensure the embedded analytics solution scales effectively with user demand.

The Sigma plugin for Claude Code is designed to streamline the process of building and managing data models directly from the terminal without requiring users to navigate the Sigma UI. It consists of two main skills: sigma-api, which handles authentication, and sigma-data-models, which instructs the agent on creating and modifying data models using Sigma's API. This plugin allows data engineers to efficiently define and update data models, including columns, metrics, relationships, and descriptions, by leveraging the agent's precise instructions for interacting with the API. The ability to manage Sigma assets programmatically within the terminal reduces the need for context-switching and accelerates the creation and maintenance of data models. This tool is especially beneficial for data engineers at companies like Plug's Electronics, enabling them to quickly build live, governed data models and incorporate schema changes without leaving their workflow. The plugin is available on GitHub and requires Claude Code version 1.0.33 or later, positioning itself as a valuable resource for engineers seeking to enhance their terminal-based data management capabilities.

The text explores the utility and functionality of AI agents, specifically through the lens of a Sigma Data Model Agent designed to streamline the process of creating and managing data models within Sigma. By automating repetitive and well-defined tasks such as schema discovery, model creation, and semantic view generation, the Sigma Agent allows data engineers to focus on the more complex and judgment-heavy aspects of data modeling. The agent requires human confirmation for major actions, ensuring that the decision-making process remains with the user while the agent handles the mechanical tasks. The key distinction between scripts and agents is highlighted, with agents possessing the ability to understand context and make informed decisions rather than simply executing predefined commands. The Sigma Agent exemplifies how AI can effectively support data engineering workflows by reducing manual effort and allowing engineers to concentrate on higher-level tasks.

Embedding analytics seamlessly into applications can enhance user experience significantly by utilizing JavaScript events, which facilitate real-time, bidirectional communication between the host application and the embedded Sigma analytics. This integration allows the host application to control the analytics display and respond to user interactions, creating a cohesive product experience rather than a disjointed one. JavaScript events operate across iframe boundaries using the postMessage API, enabling both outbound events from Sigma to notify the host of interactions, and inbound events from the host to adjust Sigma's behavior. This dynamic interplay transforms analytics from a mere visualization tool into a key product feature, allowing for customized, interactive experiences tailored to user context and actions. Security is maintained by ensuring that communications are verified and governed by existing access controls. Sigma's approach to embedding makes sophisticated analytics interactions accessible and secure, ultimately allowing analytics to integrate deeply with product workflows and user interfaces.

Automated Actions in Sigma enable the automation of recurring data workflows that traditionally require manual intervention, such as sending reports, updating tables, or triggering external system calls. These actions can now be scheduled on various cadences, including hourly, daily, or custom cron expressions, ensuring that tasks like distributing reports, refreshing metrics, and syncing data occur reliably and without human oversight. Users can configure these schedules directly within the Sigma platform's actions panel, allowing for a centralized view of all scheduled workflows. The feature enhances efficiency by converting repetitive tasks into automated processes, while maintaining security and governance protocols. Automated Actions are particularly beneficial for tasks where timing is crucial and manual processes introduce risk, such as operational monitoring, automated data operations, and system synchronization. The public beta version is available, inviting users to explore these capabilities and streamline their analytics workflows.