TechPrufer
Get started
Hex
Hex logoHex · Feature trail

Metrics Layer

6 posts · Oct 2021 – Oct 2026

What this trail covers

Hex has evolved its data analytics capabilities by integrating with dbt's Semantic Layer and Databricks Unity Catalog Metric Views, enabling users to access trusted, governed metrics through a new Metrics Cell. This allows users to specify metrics, dimensions, and time grains via a UI, generating a data frame for analysis and visualization. Additionally, Hex now supports writing dbt-flavored SQL, including refs, macros, and sources, directly within Hex SQL cells, enhancing the integration with dbt. This post also details Hex's integration with Snowflake Semantic Views, allowing users to bring Snowflake-defined models, dimensions, measures, and relationships into Hex while keeping Snowflake as the source of truth. These definitions can then be used in Threads, semantic-model exploration, notebooks, and published apps. Hex's Context Studio provides a feedback loop for data teams to observe real-world usage, identify context gaps, and validate improvements. Per-user Snowflake OAuth preserves user-level permissions and row-access/masking policies.

All Hex trails
  1. Best BI tools for Snowflake Semantic ViewsThis post details Hex's integration with Snowflake Semantic Views, positioning it as a key component of an AI analytics strategy. It explains how Hex's Semantic Model Sync brings Snowflake-defined models, dimensions, measures, and relationships into Hex, maintaining Snowflake as the source of truth. These definitions are then usable across Hex's features like Threads, exploration, notebooks, and apps. The post highlights how this integration supports self-service analytics, dashboard/app building, deeper analyst workflows, and external agent workflows via the Hex MCP server. It also emphasizes the role of Context Studio in providing a feedback loop for improving context and agent performance, and the preservation of Snowflake governance through per-user OAuth. The post contrasts this with other BI tools by focusing on the breadth of workflows supported and the ability to combine Snowflake context with other organizational context.
  2. Building semantic models in HexThis post details how to build a governed semantic model in Hex for high-stakes KPIs. It outlines a playbook for admins and data leads, emphasizing the importance of a data warehouse with gold-standard, governed data as a prerequisite. The process involves curating business questions, pairing them with Verified SQL Queries (VSQs) as ground truth, and then using Hex's semantic modeling agent to scaffold the model based on these VSQs. Key steps include checking detailed descriptions for entities, measures, and dimensions, testing the model against ground-truth queries, writing usage guides for agents, and referencing the model in workspace context files. The post stresses that semantic models are for "how much/how many" questions and should not be used for "why" questions, and that the underlying data must be clean and trusted.
  3. Are Semantic Layers the Cornerstone for AI Analytics?This post critiques the sole reliance on semantic layers for AI analytics, arguing that context is a broader and more adaptable solution. It highlights the limitations of semantic layers in handling novel questions or evolving LLM capabilities. The post advocates for a partnership approach where AI agents leverage a wider range of context, including raw data, code repositories, warehouse metadata, and business strategy documents, to provide more accurate and trusted answers. It positions Hex's approach as a more future-proof strategy for AI analytics by embracing flexibility and compounding context.
  4. What is dbt (Data Build Tool)? How It Works and Where It StopsThis post details how dbt addresses metric consistency by treating SQL transformations with software engineering discipline, including models, tests, and documentation co-located in YAML files. It explains the scope of dbt's job as a compiler and runner for SQL-based data transformations, and where its responsibilities end, particularly concerning metric drift and the bypass of governed paths. It also discusses the trade-offs between dbt Core and dbt Cloud.
  5. What is business logic? Definition and examples for data teamsThis post defines business logic as the rules and transformations that translate raw data into organizational decision-making metrics, contrasting it with business rules (the 'what'). It details common areas where business logic resides (revenue calculations, active user criteria, churn definitions) and the problems caused by its scattering across BI tools, analyst notebooks, and spreadsheets, leading to metric drift and eroding trust. It proposes a progressive centralization strategy: 1. Document and endorse tables/columns. 2. Move logic into version-controlled transformations (e.g., dbt models). 3. Formalize with semantic models. The post highlights Hex's role in this process, including semantic model sync and the Semantic Model Agent, and emphasizes explicit ownership of metric meaning vs. implementation.
  7. Semantic Modeling: How to Build a Single Source of TruthThis post details the concept and implementation of semantic modeling for data analytics. It explains the building blocks of a semantic model (entities, dimensions, measures, metrics) and how they are defined in code (e.g., YAML with dbt MetricFlow). It contrasts semantic models with raw data schemas and highlights their role in resolving metric drift and providing governed definitions for AI agents. The post outlines a practical approach to building semantic models, starting with contentious metrics, building on normalized staging models, incremental adoption, and validation through CI. It also touches upon the broader semantic context including warehouse descriptions and workspace rules, and mentions Hex's Context Studio for feedback loops.
  8. Introducing Databricks UC Metric Views integrationThis post details the integration of Databricks Unity Catalog Metric Views into Hex. It enables users to browse, query with SQL, and build data apps backed by governed metrics defined in Databricks. The integration allows for 'on-rails exploration' with trusted definitions and 'off-road' querying of raw tables with SQL and Python. Future integrations with AI and conversational analysis leveraging metric views are mentioned, as well as the ability to author metric views in Hex and push them to Databricks.
  9. Introducing new Snowflake integrations with Semantic Views and AISQLThis post details the integration of Snowflake Semantic Views into Hex, allowing users to define and ingest semantic models directly from Snowflake. It also highlights the integration of Snowflake's Cortex AISQL functions for querying unstructured data within Hex notebooks. The post emphasizes the benefits of using semantic models for AI, such as generating more accurate and helpful outputs from Hex Magic. It outlines the steps to set up Semantic Model Sync with Snowflake and details the capabilities of AISQL for analyzing images and text.
  10. Introducing Semantic Model Sync in HexIntroduced Semantic Model Sync, enabling direct integration with dbt MetricFlow and Cube semantic models. This allows users to leverage pre-defined measures, dimensions, and joins within Hex's Explore experience. The sync process involves connecting via GitHub, parsing semantic code directly, and making these governed concepts available in Explore cells and chainable into SQL/Python cells. This enhances self-serve analytics by providing trusted, governed metrics without requiring users to redefine logic within Hex.
  11. Metrics layers: past, present, futureThis post provides a historical overview and future outlook on the 'metrics layer' concept in data analytics. It details the problems with inconsistent metric definitions in the past, the rise of dbt and its approach to defining models, and the challenges faced by dedicated metrics layer startups like Transform. The acquisition of Transform by dbt Labs is presented as a strategic move to integrate a mature metrics layer (MetricFlow) with dbt's widespread adoption, aiming to solve the chicken-and-egg problem and standardize metric definitions across the data stack.
  12. Hex + dbt: Bringing trust, governance, and speed to the data stackThis post details the integration of Hex with dbt Cloud and dbt's Semantic Layer. It introduces a straightforward connection to pull dbt metadata (freshness, tests, column descriptions) into Hex's schema browser. A new Metrics Cell is introduced, allowing users to access dbt metrics without writing code. The integration also enables the use of dbt-flavored SQL (refs, macros, sources) within Hex queries.
  13. Introducing Hex’s Integration with the dbt Semantic LayerThis post introduces the integration of Hex with the dbt Semantic Layer, featuring a new Metrics Cell for accessing governed metrics via a UI and enabling the use of dbt-flavored SQL (refs, macros, sources) within Hex SQL cells. This allows users to leverage dbt's defined metrics and models directly in their Hex analyses.
  14. Getting the most out of dbt docsThis post details the integration of dbt descriptions into Hex's database schema browser. Previously, users had to navigate to a separate dbt docs site, creating friction. This new feature surfaces model, source, and column descriptions directly within Hex, alongside test statuses and a link to the dbt docs site. This aims to reduce the need for users to ask data teams for definitions and improve the discoverability and understanding of datasets.
  15. Hex 💜 dbtThis post announces the integration of Hex with dbt Cloud, allowing Hex to enrich its schema browser with metadata from dbt, including table freshness, recent job executions, and test status. This integration aims to improve observability and trust in data sources by exposing transformation layer metadata directly within the analytics workspace, reducing the need for manual data verification queries.