Articles
August 26, 2026 · 10 min read
Migrating from Mixpanel When Your Data Is Already in the Warehouse
If your events already land in Snowflake, BigQuery, Databricks, or Redshift, migrating off Mixpanel does not mean re-instrumenting your product. Here is the concrete path: what changes, what stays, and what to do with historical events.
July 31, 2026 · 8 min read
What Is an Analytics Agent? Definition and Guide
An analytics agent answers data questions by running multi-step analysis on your warehouse. See how agents work, the main types, and what makes one trustworthy.
July 31, 2026 · 11 min read
Analytics Agents vs Text-to-SQL and BI Chatbots
Text-to-SQL (NL2SQL) tools and BI chatbots have a language model write the query; analytics agents run investigations on a deterministic engine. Where text-to-SQL works, where it breaks, and a checklist for evaluating NL-to-SQL tools.
July 31, 2026 · 8 min read
Why an Analytics Agent Needs a Semantic Layer
Analytics agents are only as trustworthy as the context they run on. Why a product-analytics semantic layer prevents hallucinated SQL and methodology errors.
June 19, 2026 · 17 min read
Evaluating AI Analytics Agents for Enterprise Data: A Decision Guide for Data Leaders
A structured framework for enterprise data and analytics leaders assessing AI analytics agents: methodology trust, warehouse governance, semantic depth, diagnostic depth, security, and governance of permissions, shared definitions and audit, with a buyer checklist.
June 5, 2026 · 5 min read
CSV Upload: See Mitzu Work on Your Own Data in Minutes
Mitzu now accepts a CSV as a starting point. Drop in a file and the Configuration Agent takes over: it reads the data, identifies timestamp and event columns, maps them, and models the dataset — the same setup that normally follows a warehouse connection, handled automatically. The fastest way to see Mitzu work on something real. When you're ready for live data, connect your warehouse and everything scales from there.
June 4, 2026 · 8 min read
Mitzu's MCP Server: Product Analytics Inside the Agent You Already Use
Mitzu's analytics MCP server gives Claude, ChatGPT, Cursor and GitHub Copilot one tool: Mitzu's analytics agent. Answers come from your governed metrics, with the SQL behind them and the same permissions as in Mitzu.
June 2, 2026 · 8 min read
Custom Charts in Mitzu: When the Analytics Agent's Answer Spans Multiple Insights
Custom Charts is a new capability in Mitzu's Analytics Agent. When an answer can't be expressed by any single Mitzu insight — a DAU/MAU ratio, a Q3 vs Q4 percentage delta, two conversion rates overlaid week over week — the agent now runs the calculation across multiple insights and renders the result as an ad hoc chart inline. This guide explains what Custom Charts are, when they appear, the underlying insights you can still save, and how the feature changes the kinds of questions you can ask an AI analytics agent.
June 1, 2026 · 9 min read
Planning Mode in Mitzu: Review the AI Analyst's Plan Before It Runs the Analysis
Planning Mode is a new capability in Mitzu's Analytics Agent that converts a complex product analytics question into an explicit, reviewable plan before any query runs. This guide explains how it works, when to use it, and why showing the plan first changes the trust profile of AI analytics for data teams, PMs, and growth leads.
May 8, 2026 · 14 min read
How AI Agents Query Your Data Warehouse: From Natural Language to Product Insights
AI agents can query your data warehouse in plain English, but only reliably when governed definitions connect human language to your actual schema and the SQL is generated in a way you can check. This guide explains the architecture, the failure points, and how to turn AI insights from the warehouse into decisions.
April 23, 2026 · 11 min read
What Is Self-Service Analytics? A Guide to Governed Self-Service
What self-service analytics is, why it breaks without governance, and how to roll out a governed version where the data team defines metrics once.
March 26, 2026 · 7 min read
Can you replace data analyst with AI? A pragmatic operating model for 2026
Can AI replace a data analyst? No. Plan for AI to multiply what the data team can answer, not to cut headcount: automate repeated questions while analysts own definitions, permissions and judgment.
March 20, 2026 · 12 min read
What Is Agentic Analytics? 2026 Guide to AI Analysts and Trusted SQL
Agentic analytics explained: AI agents that answer questions, investigate why metrics moved and monitor KPIs on governed definitions, and how it differs from text-to-SQL.
March 20, 2026 · 8 min read
What is an AI data analyst? How AI analytics agents replace the ticket queue
An AI analyst, or AI data analyst, is an AI agent that answers business questions in plain language on your warehouse data, using the metric definitions and permissions your data team governs, and shows the SQL behind each answer.
March 20, 2026 · 13 min read
AI analytics hallucinations: can you trust AI analytics answers?
Can you trust AI analytics answers? Yes, when the data team owns the metric definitions, a deterministic engine rather than the language model writes the SQL, every answer shows that SQL, and the same permissions apply on every surface. How AI analytics hallucinations happen and how to keep them out of decisions.
March 20, 2026 · 10 min read
ChatGPT vs AI Analytics Agent: Which One Gives Reliable Business Answers?
ChatGPT data analysis works for uploaded files and SQL drafts. For company metrics, compare uploading CSVs, pointing an LLM at the warehouse, and an analytics MCP server on governed definitions.
February 28, 2026 · 5 min read
How Shapr3D Moved from Three Analytics Tools to One Platform
Shapr3D replaced several disconnected analytics tools with Mitzu. Product managers and marketers now build their own dashboards on one set of events, with 90% fewer data mismatches and 3x faster decisions.
February 11, 2026 · 5 min read
How Colossyan Scaled Beyond 100 People Without Growing Its Data Team
Colossyan grew past 100 people without growing its data team. Product, marketing, design and customer success build their own reports in Mitzu, ad-hoc reporting is 50% faster, and the onboarding funnel improved by 20-30%.
January 30, 2026 · 5 min read
How Mitzu Helped Prezi Reach 100% Data Accuracy
Prezi moved its product analytics off third-party tools and onto Mitzu on its own data warehouse. Reports reached 100% accuracy, data-related support tickets fell by 50%, reporting time dropped from 2 days to 2 hours, and Mitzu was 5x cheaper than the third-party tools.
November 1, 2025 · 11 min read
MRR and Net Revenue Retention (NRR): How to Calculate Both From Your Data
How to calculate MRR and net revenue retention (NRR): the formulas, worked examples, MRR movements (new, expansion, contraction, churn), NRR benchmark caveats, and a SQL sketch for subscription tables in your data warehouse.
October 24, 2025 · 5 min read
Measure Cohort Retention Without Complex SQL Queries
How to calculate cohort retention rate, read a cohort analysis table and retention curve, and measure retention on your warehouse with SQL or with Mitzu.
September 13, 2025 · 5 min read
Cruise Booking Platform Uses Mitzu for Revenue Attribution
Learn how CLICKnCRUISE uses Mitzu for affordable product analytics to improve customer retention, optimize revenue attribution, and drive business growth.
July 23, 2025 · 5 min read
Designing Analytics Stack with Real-Time Data Streaming and Event-Driven Architecture
Real-time streaming and event-driven architectures enable instant data processing and analytics, keeping insights fresh and systems responsive at any scale.
June 18, 2025 · 8 min read
The Playbook for Building a High-Impact Data Team
How to build a high-impact data team: align with the business, own the definitions and permissions, let everyone else self-serve, keep the stack lean and measure impact.
June 11, 2025 · 8 min read
Designing a Composable Product Analytics Stack with Snowplow and Mitzu
Build a flexible, AI-ready product analytics stack with Snowplow and Mitzu for precise event tracking, schema control, and powerful SQL-based insights.
June 10, 2025 · 10 min read
Row-Level Security in Analytics: Governed Data Access for Dashboards and AI Agents
What row-level security and column masking are, how they work in Snowflake, BigQuery and Databricks, and how to keep them in force for dashboards and AI agents.
May 13, 2025 · 5 min read
How 52 Entertainment's Data Team Saved 60+ Hours a Month
52 Entertainment's data team saved 60+ hours a month after product teams started answering their own questions in Mitzu, on the BigQuery and dbt stack the data team already ran. Integration from multiple data sources also got 2x faster.
May 13, 2025 · 12 min read
Delta Lake vs Apache Iceberg: Which Table Format Wins?
Delta Lake vs Apache Iceberg Which Table Format Wins comparison with features, pricing, data architecture, and which analytics stack each option fits best.
February 14, 2025 · 10 min read
Single Source of Truth for Analytics: What It Means and How to Build One
What a single source of truth for analytics means in practice, why it breaks when metric definitions are copied into each tool, how a data warehouse plus governed metric definitions delivers it, and what it does not mean.
January 4, 2025 · 11 min read
Warehouse Native vs. First-Generation Product Analytics
Warehouse-native vs first-generation product analytics: how the two architectures differ on data access, cost, governance and AI agents, why vendor-silo agents only see the vendor's copy of your data, and which stack fits which team.
December 18, 2023 · 10 min read
BI vs Product Analytics: What Each Answers and Where They Meet
BI vs product analytics: what each one answers, the data model and users behind it, where they differ, why the line blurs once both live in a warehouse, and why one platform on one set of metric definitions can answer both.