Product Updates
clariBI Forecasting: backtested projections, prediction ranges and drivers
clariBI Forecasting projects a metric forward with a method chosen by backtest on your own history, shows 50, 80 and 95…
Blog
The clariBI blog
Articles on business metrics, reporting, forecasting and getting answers from the data in the tools you already use.
Featured
Product Updates
The clariBI MCP server: use clariBI from Claude, Cursor or ChatGPT
clariBI runs its own MCP server at claribi.com/mcp/v1/. Connect Claude Desktop, Cursor, ChatGPT or your own agent and i…
Comparisons & Evaluations
Picking your first BI tool as a 10-person startup
At ten people the spreadsheet that runs the company starts to crack. Four questions decide which BI tool fits: engineer…
Latest posts
Comparisons & Evaluations
clariBI vs. Looker: when each tool is the better choice
Looker is built around a governed LookML model maintained by a data team. clariBI connects business apps directly and answers questions in plain English. Where each one fits, where each one falls short, and a quick way to decide.
Comparisons & Evaluations
clariBI vs. Tableau: an honest comparison
Tableau is a visual analytics tool for people who build dashboards by hand. clariBI connects your business apps, builds a dashboard for each and answers questions in plain English. Here is where each one is the better fit.
How-To Guides
Making the most of your 50 trial AI credits
A clariBI trial comes with 50 AI credits for 14 days. Much of what you will do in the trial costs nothing, and the rest costs more or less depending on how you ask. What uses credits, how the cost is worked out, and a plan for the two weeks.
How-To Guides
What happens when your clariBI trial ends
The trial lasts 14 days and needs no card. When it ends, your account moves to the Free plan: your data, sources and source dashboards stay, AI credits stop. Here is what changes and how to decide whether to upgrade.
How-To Guides
Your first five steps in clariBI: a trial walkthrough
You have a 14-day trial, 50 AI credits and an empty account. These five steps get you from sign-up to a first real answer from your own data, without spending many credits, plus what to leave for later.
Product Updates
Why we bet on MCP for clariBI integrations
Integration breadth decides whether a BI tool is useful to a small company. Here is why we built on the Model Context Protocol, how the catalog grew from 30 vendors to 86 apps, where MCP was the wrong fit, and what did not change.
Data Strategy
The security model behind clariBI's MCP integrations
Connecting an AI tool to Stripe or HubSpot raises fair questions. What clariBI's MCP integrations can and cannot do: read-only tools, a vetted catalog, OAuth with PKCE, encrypted credentials, per-organization scoping, and what disconnecting actually removes.
How-To Guides
OAuth 2.1 with Dynamic Client Registration, explained without the PhD
OAuth 2.1 with Dynamic Client Registration is why connecting an MCP app can be a single click instead of a trip to a developer console. Here is how it works, step by step, in both directions clariBI uses it.
Product Updates
86 apps, one catalog: inside clariBI's MCP library
clariBI's MCP catalog lists 86 apps that publish their own MCP server, from Stripe and HubSpot to Linear and PostHog. Here is what is in it, what it reads, which plan each app needs and what is missing.
AI & Machine Learning
What is MCP? The open protocol behind clariBI's integrations
MCP, the Model Context Protocol, is an open standard for connecting AI applications to other software. What it is, how it works in plain terms, what changes for the people using it, where its limits are, and the two ways clariBI uses it.