One question everysource one answer
Analytics across all your tools, for teams without a data analyst
Revenue sits in Stripe, pipeline in HubSpot, spend in Google and Meta Ads. clariBI connects them, answers questions that cross them, writes reports you can send and forecasts what comes next. The numbers are calculated with code from your own rows. No credit card for the trial.
Connect
Every tool in one place
More than 150 tools, plus files and databases. Sign in with OAuth or paste an API key. clariBI copies the data read-only, keeps the history and builds each source its own dashboard with no AI and no credits.
Combine
Questions that cross your tools
Ask how Google and Meta ad spend relates to new Stripe revenue. clariBI lines the sources up by date or a shared column, compares and correlates them, and returns a chart, the table and a short explanation. The maths is done in code over your rows.
Report
Written reports you can send
An executive summary, findings and risks, with the charts they refer to, written from all the sources you pick. Every number in the text is checked against the chart data. Download as PDF, or schedule it by email on Professional and Enterprise.
Forecast
Forecasts with honest ranges
clariBI tests several methods on your history, keeps the one that would have predicted best and shows 50, 80 and 95 percent ranges. It also checks which of your other metrics move with it. No language model does the maths.
One question, three sources
Ask it the way you would ask a colleague
clariBI finds the sources that hold the answer, here Google Ads, Meta Ads and Stripe, lines them up by week and calculates the figures from the synced rows. The example uses sample data.
Weekly spend stayed between $4.3k and $5.3k. New revenue rose in 8 of 11 weeks, from $18.2k to $31.2k, so each dollar of spend brought in about $6.40 by the end, up from $4.20.
Marketing and revenue, Q3
Paid acquisition review
Executive summary
New revenue rose from $18.2k to $31.2k a week while ad spend stayed between $4.3k and $5.3k. Each dollar of spend brought in about $6.40 by the end of the quarter, up from $4.20.
Findings
- Revenue grew in 8 of the 11 weeks. Each of the 3 weeks that dipped was followed by a new high.
- Spend changed little, so the gain came from more revenue per dollar, not from a bigger budget.
Risks
- The data shows spend and revenue side by side. It does not show that the ads caused the growth.
From answer to report
A report your team can read
Save any answer as a report, or start from a report template. The language model writes the summary, findings and risks. Each figure has to match the chart data, or the sentence is removed. Same sample data as above.
Every source also gets its own dashboard
A real screen from clariBI, built from a sample CSV of daily store sessions, orders, revenue and ad spend. Nobody arranged these widgets by hand, and building them used no AI credits.
Data freshness. Every dashboard says how recent its data is and when it last synced, with Sync now.
Totals with context. Each total shows its change against the previous period, plus average and median.
Next steps. Generate a written report, forecast a metric or share the dashboard from the same page.
Also in clariBI
Works with the tools you already pay for
What the AI does, and what it does not
- Plan how to answer your question and pick the chart
- Write the explanation of results that code calculated
- Propose dashboard widgets and write report text
- Change anything in your tools; every connection only reads
- Do the maths: totals, averages, trends and forecasts are calculated with code
- Keep sentences in AI reports whose numbers do not match the chart data
AI work uses credits and you can see your balance in the app. The Free and Lite plans do not use AI at all.
Plans from $0
The 14-day trial needs no credit card and includes 50 AI credits, 3 data sources and 1 GB of storage. When it ends, your account moves to the Free plan: your data and dashboards stay, AI features stop until you choose a plan.
| Plan | Price | Users | Data sources | AI credits / month |
|---|---|---|---|---|
| Free | $0 | 1 | 3 | None |
| Lite | $19 / month | 1 | 5 | None |
| Starter | $99 / month | 3 | 10 | 500 |
| Professional | $199 / month | 15 | 50 | 1,500 |
| Enterprise | $999 / month | 100 | 100 | 5,000 |
For developers
Use clariBI from Claude, Cursor or your own code
The MCP server gives AI assistants 26 tools: list dashboards and reports, upload data, run analyses and forecasts. A REST API with an OpenAPI reference is available on Professional and Enterprise.
{
"mcpServers": {
"claribi": {
"url": "https://claribi.com/mcp/v1/",
"headers": { "Authorization": "Bearer claribi_mcp_..." }
}
}
}
# or: pip install claribi-mcp / npm install @claribicom/mcp
From the blog and the help center
Recent posts
- clariBI Forecasting: backtested projections, prediction ranges and drivers
- The clariBI MCP server: use clariBI from Claude, Cursor or ChatGPT
- Self-service BI vs. embedded analytics: how to choose
- Migrating from Excel to a real BI tool
- clariBI vs. Power BI: which fits your stack
- Picking your first BI tool as a 10-person startup
- Metabase vs. clariBI: a fair look at when each fits
- Dashboard vs. report: when to use which
Help center
- MCP Server Integrations
- Manual Data Import Guides
- Understanding the Home dashboard: every section explained
- Getting Started with clariBI: Your First Dashboard
- Data source errors and how to fix them
- Dashboard builder reference: metrics, layout, refresh and drill-down
- The 24 widget types and when to use each one
- 30 example questions that get good results
Try it on your data
Connect one tool, ask one question and see whether the answer is useful. Fourteen days, no credit card.