Structured (YC S23)

Metrics you can fall back on 🍁

🎃 October Product Update 🕸️ New Feature: StructuredBot PR Generation This October, Structured conjures up automated cleanup pull requests to keep your data from turning into a nightmare! StructuredBot is here to scare off inconsistencies and help you keep your metrics eerily clean and consistent. Here’s how it works: Summon StructuredBot via GitHub Issues 1️⃣ Invite StructuredBot into Your GitHub Repository 👹 Start by inviting StructuredBot into your GitHub repository, allowing it access to make changes. Once installed, "summon" the bot by tagging an issue with “cleanup.” This triggers StructuredBot to begin its work. 2️⃣ Mark Issues with the “Cleanup” Spell 🧙 Once StructuredBot detects the tag, it scans your repository's indexed data models, searching for data inconsistencies and optimization opportunities based on predefined rules and heuristics. 3️⃣ Let StructuredBot Do the Witchcraft 🔮 StructuredBot scours your indexed data models to detect and correct haunted data, generating a cleanup PR. It identifies issues like missing documentation, redundant models, inconsistent calculations, and inefficient joins. StructuredBot enforces naming conventions, standardizes calculations, and tests for data integrity to keep your metrics clean and spooky-free—all within a single cleanup PR. 4️⃣ Review, Approve, and Exorcise Your Metrics 💀 After analyzing and optimizing, StructuredBot generates a PR consolidating all changes, providing a clear record of each update. Review the bot’s recommendations, and if satisfied, approve and merge the PR. This final step “exorcises” inconsistencies, leaving your data clean and ready for accurate analytics. 🕷️

TechCrunch Disrupt has been an incredible experience so far! 🚀 Can’t wait to hit the stage in just 2 hours

Get set up in 5 Minutes or less — Book a demo today. https://lnkd.in/djUu2faN

Get set up in 5 Minutes or less — Book a demo today. https://lnkd.in/djUu2faN

Get set up in 5 Minutes or less — Book a demo today. https://lnkd.in/djUu2faN

Our first #TechCrunchDisrupt2024 is just around the corner, and we can't wait! 💡 Catch us at Startup Battlefield on Oct 30 🎁 Swing by our booth for live demos and exclusive swag Looking forward to see you there!

Structured CTO in founder mode

Come see our CEO & co-founder speak at GHC in Philly next week -- Amrutha Gujjar 💯

🔧 dbt Labs is great for transforming raw data from your warehouse (Snowflake, BigQuery, etc.) into analytics-ready models. But its flexibility can lead to fragmentation. Different teams might define the same metrics—like revenue or customer lifetime value—in slightly different ways. Over time, these small differences add up, creating misaligned definitions and making it impossible to trust anything as a single source of truth. 🔍 Our GitHub app fixes this by addressing the problem proactively at the pull request level—before these issues become company-wide headaches. Here’s how it works: - 1️⃣ When engineers submit a PR for a new DBT model, the app checks if similar logic already exists. If Product builds a revenue model, the app flags the PR and prompts them to reuse Finance’s version. This ensures all teams pull from the same source of truth. - 2️⃣ If a PR introduces a metric (like churn) that conflicts with your company’s semantic layer, the app flags it right away. This keeps Finance, Marketing, and Product aligned on key metrics. - 3️⃣ Engineers often rebuild models that already exist. Our app detects redundant efforts in PRs, suggests reusing existing models, and saves hours of work. This makes your DBT project more modular and scalable. The challenge isn’t DBT itself—it’s how teams are applying it. As your company scales, you need modular, consistent data models. By automating PR reviews, this app helps unify your models, reduce errors, save time, and ensure your entire company works from the same data. 🔗 Try it now on GitHub Marketplace (Link in comments). Structured (YC S23) #DBT #DataEngineering #AI #RevenueMetrics #ModernDataStack #DataGovernance