October 2022 Summaries

Normalization is crucial for the Extract-Load then Transform (ELT) process as it ensures correctness, flexibility, and understandability of data. Traditionally, data analysts accessed modeled data in a data mart due to technology limitations. However, with advancements in cloud data platform architectures, data modeling can be separated from transformation and analysis by landing the data in a normalized schema. Normalization allows for verifiable correctness, flexibility to changing downstream needs, and intuitive understandability by data modelers. Fivetran delivers normalized schemas that are as close to third normal form (3NF) as possible, automating the Extract-Load process into a normalized schema and freeing up valuable data engineering time for higher-valued activities.

Navigating the process of unloading data from Snowflake involves understanding the various methods and tools available, such as the COPY INTO command, which allows users to transfer data to different destinations like Amazon S3, Google Cloud Storage, or Microsoft Azure. The choice between manual and automated data exportation depends heavily on the intended use of the data, whether for integration with business tools or migration to other data services. Ensuring data backup is crucial to prevent data loss during manipulation, and utilizing file types like Parquet can optimize storage, especially for large datasets. Users should be mindful of setting up necessary integrations and considering file formats compatible with their destination. For those seeking a more streamlined approach, Fivetran Activations offers a reverse ETL solution that simplifies transferring data from Snowflake to over 85 destinations, providing enhanced speed and observability without complex setup.

Fivetran has expanded its partner program and introduced new certifications to reward partners for their business while providing additional sales, enablement, and support opportunities. The company launched a referral program in 2018 and has since collaborated with hundreds of partners globally. The newly expanded certification program educates and empowers partners to take the new offerings to market and support customers effectively. Fivetran now offers four tiers for its referral program, three tiers for its reseller program, and four tiers for its training and certification program. The company is actively working on making updates to its partner onboarding journey and will continue updating the program to provide partners with the best experience possible.

The adoption of a modern marketing data stack is helping organizations improve the quality of their marketing analytics at less cost, leading to increased overall marketing efficiency. By employing a best-of-breed approach instead of building solutions from scratch, companies can centralize marketing data and gain a 360-degree view of customer behavior. This enables them to drive massive efficiencies in marketing spend and targeting, as well as save engineering costs and generate more business value. The combination of Fivetran and Snowflake has been instrumental in helping organizations achieve these benefits.

Fivetran has released new data connectors and data models for its customers to access and transform data more easily. The company's beta releases are available to all users with a Fivetran account, while private preview versions are only accessible by a select group of customers. Connectors in beta include those for Oracle, NetSuite, SAP Business ByDesign, ADP Workforce Now, Amazon Ads, Google Analytics 4 Export, and more. Meanwhile, connectors in private preview include those for SQL Server, Snowflake as a Source, Elasticsearch, OpenSearch, The Trade Desk, RingCentral, Medallia, Oracle Cloud Fusion App Suite: ERP, CX, HCM, SCM, and others. Upcoming connectors are also listed, including High-Volume Agent DB2 for i, BigQuery as a Source, PayPal, Microsoft Dynamics 365 Business Central, Freshservice, Bamboo HR, Amazon Seller & Vendor Central, SalesLoft, Google Search Ads 360, Spotify Ads, Yahoo DSP, and more. Additionally, Fivetran has updated its data models for marketing analytics, finance analytics, operations analytics, and other areas.

Fivetran has announced its 2022 global Partner Award winners, recognizing top ecosystem partners for their performance and impact in delivering demonstrable customer outcomes and business impact. The awards were presented at the company's Partner Summit on October 20th. Key categories included Global Tech Alliance Awards, Business Intelligence Partner of the Year, Ecosystem Partner of the Year, and SI Partner Awards across multiple regions. These partners have demonstrated outstanding results in helping organizations implement and scale modern data stack solutions and Fivetran to drive superior business outcomes, operational efficiencies, and customer service.

The webinar on selecting the right data integration tool for Revenue Operations (RevOps) discusses the benefits and differences between Integration Platform as a Service (iPaaS) and Reverse ETL. Experts Ben Lewinsky and Sylvain Giuliani highlight that iPaaS tools create direct connections between apps, whereas Reverse ETL syncs data from a single source of truth, such as an analytics warehouse, to various business tools, promoting reliable and consistent data use. The importance of strategic data planning, the value of SQL skills in RevOps, and the role of RevOps in aligning sales, marketing, and customer success teams to drive revenue are emphasized. The concept of "star metrics" is introduced to unify teams and guide product development, and the necessity of starting small and evolving analytics strategically as an organization grows is advised. Ben underscores the critical nature of open communication and prioritization across departments to ensure effective RevOps operations.

Fivetran, a data integration service, has become a Metrics Ready Partner for dbt Semantic Layer. This partnership allows users to enhance their data models with rich metric definitions and derive value from their data more quickly and easily. Fivetran's first dbt metrics package extends Ad Reporting data models with spend, engagements, and clicks metrics. The integration also provides secondary calculations for period-over-period comparisons, period-to-date aggregates, rolling calculations, and custom metric creation. These predefined business metrics are hosted on a Git project, enabling version control and collaboration.

Integration Platform as a Service (iPaaS) tools, such as Zapier, Workato, and Boomi, offer an accessible, no-code interface for connecting software applications, making them popular among non-technical business units for streamlining workflows and avoiding repetitive data entry. However, these tools are not designed to handle large-scale data integration, as they can become unwieldy and error-prone when used beyond their intended point-to-point connections. This often results in data integration issues, lack of visibility, and compromised data quality. As companies grow, they may need to transition to Reverse ETL solutions, which provide a hub-and-spoke model for data management, ensuring that all applications have access to a single, reliable source of truth from the data warehouse. Reverse ETL allows for batch-based transformations and declarative data syncing, offering better scalability, security, and maintenance compared to iPaaS, which becomes inefficient and costly due to its imperative, trigger-based method. Ultimately, businesses should employ the right tools for their data management needs, potentially using both iPaaS and Reverse ETL to maintain effective data pipelines and ensure data reliability, especially as they scale.

The Fivetran Wizard for dbt Core™ and Google Cloud BigQuery are tools designed to enhance data transformation processes within modern data stacks, addressing the challenge many companies face in transforming raw data into actionable insights. Fivetran facilitates automated and reliable data ingestion using change data capture technology, seamlessly moving data to cloud destinations. BigQuery, a fully managed enterprise data warehouse, offers advanced data management and analytics capabilities, making it ideal for processing complex data transformations. dbt empowers data teams to perform scalable data modeling using SQL, incorporating engineering best practices such as reusable macros, version control, and auto-documentation. To further streamline modeling, Fivetran’s Wizard for dbt Core™ extension for Visual Studio Code provides functionalities like auto-completion, error highlighting, SQL to ref conversion, and easy installation of dbt packages, making it easier for analysts to write efficient models and reduce the time spent on basic model foundations. Together, these technologies enable organizations to focus more on data-driven decision-making and less on the intricacies of data pipeline management.

Fivetran has partnered with Microsoft for the Intelligent Data Platform Partner Ecosystem, allowing enterprise customers to implement fully automated, managed, no-code data pipelines for reliable and secure data delivery on Azure. This partnership accelerates data movement to Azure Synapse, enabling organizations to democratize analytics across their organization more responsibly. Fivetran's integration with the open and well-governed Intelligent Data Platform simplifies and automates the ELT motion, offering simple ways to clean and deduplicate data, normalize it, block and hash columns with PII, and more.

Product-led growth (PLG) has the potential to revolutionize Revenue Operations (RevOps) by utilizing data ecosystems to drive more impactful outcomes, though it comes with its own set of challenges. Insights from a recent webinar featuring RevOps leaders, including Henry Mizel from Apollo and Cyril Marques from Montreal Analytics, emphasize the importance of integrating product data into operational strategies to enhance customer engagement and conversion rates. By centralizing data in a warehouse, companies like Apollo have leveraged product usage insights to prioritize leads, automate upsell alerts, and improve outbound conversion rates. The process involves ensuring data reliability and creating a unified source of truth, which allows for more authentic customer interactions and better-informed sales strategies. The webinar highlighted the necessity of adopting an agile approach to data activation, focusing on building solid foundations and iterating over time to refine PQL scoring and other metrics. This approach not only aids in personalizing customer interactions but also supports continuous experimentation and adaptation to meet business goals effectively.

An Enterprise Data Warehouse (EDW) is a centralized repository that consolidates and standardizes data from all areas of a business, enabling departments to share information easily and improve decision-making. It differs from a regular data warehouse in its size and scope, as it supports multiple departments within an organization with a unified approach to organizing and representing data across various processes and teams. There are three types of EDWs: on-premises, cloud, and hybrid. The architecture of an EDW typically follows a three-tier design, consisting of the data repository, OLAP server, and front-end layer. Benefits of using an EDW include immediate access to actionable data, multi-department collaboration, consolidated and standardized data, empowerment of less technical team members, and adherence to compliance requirements. When evaluating potential vendors for an EDW, consider factors such as compatibility with existing systems, cost structures, scalability, security features, user access controls, fault tolerance, and peer recommendations.

In exploring the revenue dynamics of fast-growing companies, George Fraser analyzes the intriguing behavior of businesses with varying growth and net revenue retention (NRR) rates, using Fivetran as a case study. He illustrates how a company that doubles its customer base annually and sees a 50% growth in existing customers can initially experience a 150% growth rate, which over time converges to the customer growth rate, eventually stabilizing around 100%. This phenomenon occurs because the proportion of growth from new customers diminishes as overall revenue increases, which means the long-term growth rate aligns with the customer growth rate rather than expansion. Despite the initial impression that positive NRR might indicate a higher growth rate, it actually equates to higher average customer value, as long as the customer growth rate surpasses the expansion rate. The study highlights that while NRR enhances a company's value and revenue, the fundamental driver of growth remains the number of new customers.

N26, a neo-bank from Germany, has successfully integrated the Data Build Tool (dbt) into its data architecture, significantly enhancing the productivity and scope of its Data Analytics team. Initially, N26 faced challenges related to data governance and pipeline management, relying on in-house ETL solutions that required analysts to code using Python and SQL. However, with dbt, which supports SQL-based data transformation pipelines, N26 streamlined its data processes, allowing analysts to focus on building and maintaining core data models without needing extensive SQL admin expertise. The adoption of dbt has led to the creation of complex data projects, with over 800 models running regularly, and has fostered a more autonomous data analytics environment. Despite some challenges related to data architecture knowledge and dependency management, dbt has proven integral to N26's data strategy, prompting plans to hire analytics engineers and explore tools like Airflow to further improve workflow orchestration and data quality.

Fivetran has introduced SAP AppConnect, a solution that simplifies data integration for enterprises using SAP applications. The tool enables log-based change data capture despite common license restrictions for databases managed by SAP applications. SAP ERP data is central to many companies' business processes, and accessing, combining, and analyzing this data can help businesses make more informed decisions. However, high-quality enterprise data is often locked away due to SAP license restrictions and complex data integration solutions that risk business operations. Fivetran's SAP AppConnect optimizes the initial load during the data replication process and simplifies data integration to ensure comprehensive visibility and access without any business disruptions.

A database is an organized collection of data stored in a computer system. It plays a vital role in handling mission-critical data for companies, including building applications, recording business transactions, managing customer data, and supporting business intelligence. There are two main types of databases: relational and non-relational. Relational databases use SQL to manage data and perform queries, while non-relational databases store data in a non-tabular format. The most popular database management systems include Oracle Database, MySQL, Microsoft SQL Server, PostgreSQL, and MongoDB.