Skip to content
EN
English 简体中文 soon 日本語 soon

Analytics Model

Ask business questions in plain language and get charts back

Visit official site

What Analytics Model is

A conversational analytics layer. Connect your data sources, then ask questions in ordinary language and get answers, charts and insights back instead of writing queries or waiting on a report.

It is also embeddable, so product teams can put the same capability inside their own application.

What you can do with it

  • Ask for a metric, comparison or breakdown and get a visualisation
  • Set smart alerts when something moves
  • Give different roles different views of the same data
  • Embed analytics into a product or internal platform through the API

Who it is for

  • Business analysts and management reporting
  • Product teams embedding analytics
  • Non-technical users who want direct access to numbers

What to watch out for

  • Natural language analytics is only as trustworthy as the metric definitions behind it
  • Agree calibres, handle sensitive fields and audit key metrics before anyone acts on output
  • A fluent explanation of wrong data is worse than no answer
  • Complex metrics still need a data team to define them, and access control has to be set deliberately

Pros & cons

✓ What we like

  • No SQL needed for common questions
  • Broad data source connectivity
  • Alerts and role-based views
  • Embeddable into products via API

! What to watch out for

  • Depends entirely on metric definitions and data quality
  • Complex metrics still need data team involvement

FAQ

Do I need SQL?

Not for routine questions, though complex metrics still need a data team to define how they are calculated.

Can it be embedded?

Yes. There is an embedded API for putting analytics into a product or internal platform.

Can insights drive decisions directly?

Use them as reference. Check data sources, time frames, definitions and outliers before acting on anything material.

Last reviewed: 2026-09-15

More AI office assistant tools

View all →

How we review