October 2025 Summaries

Census, a Fivetran company, is expanding its capabilities to integrate artificial intelligence into data workflows by introducing two new features: Embedding Columns and new vector store connections. These innovations allow data teams to generate vector embeddings directly within datasets using models like OpenAI or Gemini, and sync them to vector databases such as Pinecone and turbopuffer, facilitating AI-ready data without the need for custom pipelines. This advancement addresses the complexity of preparing data for AI, streamlining the process by embedding it into existing Census workflows, making it easier to transform structured data into vector embeddings for use with large language models and retrieval systems. An example of this is enhancing support ticket searches, where embeddings enable the identification of semantically similar issues despite differences in phrasing, thereby improving issue resolution and support automation. These new capabilities leverage existing infrastructure, ensuring data freshness and flexibility, and positioning the data warehouse as a foundation for AI-driven applications, including CRM intelligence and intelligent support systems.

NetSuite, a powerful but complex ERP system, is challenging to integrate, and Fivetran simplifies this process with its NetSuite connector, which supports both current and legacy NetSuite platforms. This connector efficiently handles a wide array of data types and ensures automation, scalability, and reliability, while offering predictable pricing by charging only for changed data. Fivetran's solution addresses the complexities of NetSuite's data model and extraction methods, which often involve cumbersome processes due to inconsistent APIs and documentation. By managing schema drift and providing a dedicated dbt package for data transformation, Fivetran enables companies like Grafana Labs and Parachute Home to streamline their data integration processes, significantly reducing development time and enhancing analytics capabilities without the need for extensive data engineering resources. This allows businesses to focus on deriving insights and making informed decisions rather than dealing with infrastructure challenges.

Fivetran Professional Services offers a comprehensive approach to optimizing data operations by providing tailored guidance, best practices, and efficient architectures for data leaders across various organizations. These services enhance the value of automated data movement, centralize and transform data with ease, and enable seamless integration with existing workflows. By leveraging their deep connection with Fivetran's product and development teams, the service ensures faster insights, lower total cost of ownership, and secure connectivity through strategies like Private Link and VPNs. The Professional Services team provides ongoing support beyond deployment, focusing on automation, scalability, and governance, which helps organizations handle large volumes of data without increasing complexity. This results in robust, future-ready data infrastructure that accelerates analytics and AI initiatives, evidenced by success stories in industries such as logistics, healthcare, and AI. Through collaborative workshops and design sessions, Fivetran empowers teams with the expertise to manage and evolve their data strategies, aligning deployment with compliance and operational priorities, ultimately driving measurable impact and innovation.

Databricks Apps offer a powerful, yet often underutilized, feature that allows users to build production-ready data applications using popular web frameworks like Streamlit, Gradio, Shiny, Flask, Dash, and Node.js within the Databricks environment. This feature integrates with Databricks' security and governance capabilities and can be used to create various applications such as reports, dashboards, predictive analytics, and GenAI applications. Despite its potential, the initial setup of Databricks Apps was time-consuming, requiring repetitive manual configurations. To streamline this process, an automation script has been developed, eliminating the need for manual setup by automatically configuring resources, assigning permissions, and deploying a functional app within minutes. The script supports all compatible frameworks, simplifying the transition from naming an app to having a tested, deployable application ready for customization. This automation aligns with Fivetran's philosophy of enhancing ease of use and scalability, enabling data teams to focus on building impactful analytics solutions by reducing the setup workload.

Fivetran offers a usage-based pricing model centered on Monthly Active Rows (MAR), which focuses on the distinct primary keys added or updated in a source and synced to a destination each calendar month, excluding deletions. This approach ensures predictable costs tied to data changes, with charges applied once per month regardless of the number of updates. Fivetran's data movement is facilitated through connectors tailored for specific source-destination pairs, with costs influenced by the volume of data moved, the plan chosen, and commitment to annual contracts. Additionally, Fivetran's transformation costs are based on successful monthly model runs (MMR), with the first 5,000 runs free and a decreasing per-run rate for higher volumes. The platform helps reduce costs through various free offerings, such as initial and historical syncs for new connections and schema-driven re-syncs. Fivetran positions itself as a fully managed data movement platform, emphasizing its ability to automate and maintain data pipelines, thereby freeing resources for high-impact work and providing an alternative to building custom data architectures. It offers significant features like auto schema drift handling, change data capture, prebuilt connectors, and enterprise-grade compliance, differentiating it from traditional ETL tools.

Fivetran utilizes a metric called revenue-weighted feature usage (RWFU) to assess the effectiveness of its product features, providing insights beyond raw adoption rates by considering the revenue generated by accounts actively using a feature. This metric helps the company identify which features resonate with their highest-value customers, guiding investment decisions. RWFU is operationalized as a standard part of Fivetran's product evaluation, integrated into dashboards and roadmap reviews, and is complemented by traditional metrics such as account-weighted usage and engagement. By combining these metrics, Fivetran gains a comprehensive understanding of product adoption, particularly useful in scenarios involving early-stage features, feature deprecation, and setting adoption goals tied to revenue impact. The company emphasizes the importance of having a robust data infrastructure to reliably apply RWFU, using its own platform to centralize telemetry and make the metric scalable.

Fivetran and dbt Labs are merging with the vision of creating an independent company focused on openness and interoperability, aiming to establish a comprehensive data platform for diverse workloads over the next decade. Emphasizing their commitment to maintaining dbt's open-source nature, the merger seeks to future-proof their offerings to accommodate evolving technologies, such as AI, by ensuring compatibility with the entire data ecosystem. The priority is to facilitate customer adoption of data lakes using open table formats like Iceberg, enabling centralized data management for varied workloads. Recognizing dbt's pivotal role in transforming analysts' careers and its significance within the data ecosystem, the merged entity is dedicated to enhancing dbt while maintaining its role as a crucial tool for data practitioners.

Fivetran and dbt Labs are joining forces to set the standard for open data infrastructure, combining the reliability of managed infrastructure with the flexibility of open standards. This partnership aims to address data siloes and enable AI to move from experimentation to execution by providing an open, interoperable foundation that decouples storage from compute. With over 1,500 joint commercial customers and 16,000 dbt projects run weekly by Fivetran, the collaboration capitalizes on Fivetran's expertise in automated data movement and dbt's analytics engineering capabilities. Both companies are committed to maintaining open-source and source-available licenses for their technologies, ensuring that openness, portability, and interoperability remain central to their mission. As AI reshapes how organizations use data, Fivetran and dbt's unified platform promises faster decisions, reduced integration bottlenecks, and scalable data systems, providing enterprises with the tools to harness AI's potential effectively.