May 2025 Summaries

Hex is a company that focuses on building a data tool to address the pain points of fragmented and outdated analytics workflows. The company's primary goal is to redefine analytics workflows in the AI era, focusing on creating a platform that captures the virtuous cycle of data work, including exploratory data analysis, canonizing and curating results, self-serve exploration, and providing valuable context for AI. With over 1,500 teams worldwide using Hex, the company has achieved significant success, but recognizes that it is still just scratching the surface of what is possible. Recently, Hex raised $70M in funding from a group of investors who are familiar with the company's vision and are looking to support its continued growth and development. The company is now hiring and invites users to get started with Hex or schedule time with one of its team members, while also announcing its presence at Snowflake Summit this year.

Huckberry, an e-commerce company with over 100,000 SKUs, was struggling to accurately forecast demand across its vast product range. The company's marketing and planning teams were at odds over inventory levels, and the use of legacy BI tools couldn't provide the granular insights needed. Huckberry's data team, led by Ben Saxe, built an interactive forecasting app using Hex that allows business users to run scenario modeling, forecast at the SKU level, and zero in on what matters most - profit. The app eliminates blind spots, enables accurate predictions up to 26 weeks ahead, and provides real-time updates as trends change. With this new solution, Huckberry has achieved detailed, dynamic forecasting, unlocked over $1M in savings, and freed up time and capital for the company's growth strategy.

How embedded analytics should feel like part of your product, not a bolted-on component, and how Hex makes it easier to build seamless experiences with flexible APIs, white-label theming, and a workspace built for speed and collaboration. With Hex, teams can launch embedded analytics without requiring a deep integration sprint or new roadmap, building their app in Hex using SQL, Python, or no-code blocks, and then generating a secure, single-use URL that drops into the product with a simple API call. This platform enables self-service analytics through simple inputs, filters, and variable controls, allowing teams to explore curated data sets on their own without needing help from a data analyst.

Companies use data workflows to transform raw data into insights. A well-designed data workflow reduces errors and saves time, as it is scalable and repeatable, allowing for the analysis of different datasets in a specific order. Data workflows are composed of various stages, including ingestion, processing, transformation, analysis, visualization, and governance, each with its own set of tools and techniques to manage data quality, security, and scalability. Industries such as healthcare, credit card transactions, retail, and construction rely on data workflows to automate tasks, improve efficiency, and drive business insights. To optimize data workflows, it is essential to define clear objectives, choose the right tools for each stage, create well-documented workflows, prioritize data quality checks, and implement orchestration, ETL, and monitoring tools. By doing so, companies can increase team efficiency, reduce costs, improve business processes, enhance collaboration, minimize downtime, and leverage emerging technologies such as AI-powered workflows and self-healing systems to drive real-time decision-making.

Data collaboration is about building a culture where teams work together to solve real problems using their datasets and expertise, rather than just sharing numbers or commenting in a doc. It involves setting up the culture, context, and workflows for meaningful contributions, even across time zones. Data collaboration is proactive, intentional, and dynamic, aiming to unlock insights that drive action without endless meetings or PDF ping-pong. It requires tools that facilitate real-time collaboration, flexible sharing, drag-and-drop app building, comment threads, multimodal workflows, API support, data privacy, and security. By implementing these elements, teams can create better workflows, make better decisions, and achieve their biggest business goals.

Self-service analytics is about empowering users to explore and visualize data without relying on a data team, reducing bottlenecks, and fostering a data-driven culture. A self-service analytics platform should support both technical and non-technical users, providing curated dashboards, drill-down capabilities, and exploration tools. When done well, it fosters faster decision-making, healthier data literacy, better resource utilization, and an empowered organization. However, poor connection to data sources, missing or inconsistent data models, defaulting to BI tools that aren't built for exploration, no plan for data governance, skipping documentation and data lineage, and not supporting advanced workflows can lead to chaos. Foundational practices include building clean, reusable data models, integrating trusted data sources, centralizing access, promoting user experience, creating clear documentation, and making space for iteration. Intentional implementation of features like AI-powered tools can streamline analysis, reduce repetitive tasks, and surface insights faster without compromising the rigor of workflows. Ultimately, self-service analytics aims to make analytics part of everyone's daily workflow, empowering stakeholders while giving data teams the space to focus on strategy, modeling, and proactive insight generation.