INTELLIGENCE
Archive
How to Set Up Hightouch for Reverse ETL
Hightouch is a reverse ETL platform that reads data from your data warehouse or database and syncs it to downstream business tools like Salesforce, HubSpot, Braze, and Google Ads. Unlike forward ETL, which moves raw data into a warehouse, reverse ETL pushes transformed, model-ready data back into the tools your sales, marketing, and ops teams actually use. Setup takes about 30 minutes for a basic source-to-destination sync.
How to Get Started with Polars for Data Analysis
Polars is a DataFrame library written in Rust that processes tabular data 10 to 30 times faster than pandas for most operations. To get started, install it with pip install polars, load a CSV with pl.read_csv(), and use its expression-based API to filter, group, and aggregate data. Polars handles large files that cause pandas to slow or crash, with no extra configuration needed, making it a practical step up for analysts who regularly hit pandas performance limits.
How to Set Up Grafana for Business KPI Monitoring
Grafana is an open-source analytics platform used by more than 20 million people worldwide, best known for infrastructure monitoring but equally capable of tracking business KPIs. By connecting Grafana Cloud to a PostgreSQL or MySQL database and writing a few SQL queries, operations managers and analysts can build live dashboards for revenue, conversion rates, and churn without installing any server software. The free tier supports up to three active users and is enough to cover most small-team reporting needs.
How to Build an Internal Dashboard with Retool
Retool is a low-code platform that lets you connect a database, API, or Google Sheets and build a working internal dashboard in under 30 minutes. You create queries against your data source, drop charts and tables onto a canvas, and share the result with your team via a URL. No backend code is required. Retool supports PostgreSQL, MySQL, Supabase, REST APIs, Google Sheets, and more than 50 other data sources out of the box.
How to Connect Your Database to Claude with MCP
Model Context Protocol (MCP) connects a PostgreSQL, MySQL, or SQLite database directly to Claude so you can query it in plain English without writing SQL. Setup takes under 10 minutes: install a lightweight MCP server, add a connection string to Claude Desktop's configuration file, and Claude reads your live schema automatically. By March 2026, over 10,000 MCP servers existed across public registries, making this the fastest path to AI-powered database analysis without custom code.
How to Query Your Database with Vanna AI
Vanna AI is an open-source text-to-SQL framework that translates plain-English questions into SQL queries against your database. After connecting to PostgreSQL, SQLite, Snowflake, or another supported database and training Vanna on your schema with example question-to-SQL pairs, anyone on your team can get answers without writing SQL. Vanna 2.0, released in early 2026, adds a built-in web interface suited for team deployment and streaming responses.
How to Use Airtable AI for Data Analysis
Airtable's built-in AI features let you classify, summarize, and extract insights from structured data without writing formulas or SQL. Using the AI field, categorization templates, and the Omni assistant, non-technical teams can turn raw records into analysis-ready tables in minutes. The workflow covers CRM data, customer feedback, survey results, or any table with more than a few hundred rows. No plugins or external tools are required.
How to Build Data Reports with Evidence.dev
Evidence.dev is an open-source framework that lets analysts build published data reports by writing SQL queries inside Markdown files. Instead of configuring a drag-and-drop BI tool, you define queries in code and Evidence renders them as a live website with charts and tables. Setup takes under 30 minutes for a CSV-based project. Evidence supports DuckDB, PostgreSQL, BigQuery, Snowflake, and flat files, producing a static site deployable to Netlify, Vercel, or GitHub Pages with full Git version control.
How to Automate Weekly Data Reports with Make
Make is a no-code automation platform that lets ops managers and founders build scheduled data pipelines without writing code. To automate a weekly report, you create a scenario with a schedule trigger, pull data from sources like Google Sheets, Airtable, or a CRM, use a text aggregator to format the output, then route it to email or Slack. Setup takes under 30 minutes and runs reliably on its own after that.