Claude Code · Anthropic
Let leaders ask business metric questions and get sourced answers
Anthropic's data team describes how it answers most internal business analytics questions with Claude Code, using skills that define how each headline metric is calculated, with human sign-off on anything going to leadership.
What they did
Each data domain has skills, folders of Markdown that Claude Code reads on demand, describing tables, definitions and how headline KPIs are calculated. A standing check compares each domain's top KPIs daily against the approved dashboard, and every answer carries a footer with its source, data freshness and owner. Skill changes are now part of most data model updates.
How it went
The company says about 95% of its business analytics queries are now handled this way at roughly 95% accuracy overall, versus no more than 21% on its tests before skills were added.
Worth knowing
Accuracy drifted from about 95% to about 65% in a month when skills were not maintained, and the team says plausible but wrong answers are not fully solved. Leadership-bound answers still need explicit human sign-off.
Try it with Claude Code
Read the metric definitions in <your metrics skill folder> before answering. Then answer: <your business question>, using only the tables named there. Show the query you ran, the result, and a footer with the data source, how fresh the data is and who owns the metric. If the definition is missing or ambiguous, stop and tell me instead of guessing.
We wrote this prompt from the case. Replace the parts in angle brackets with your own details.
What this label means: Vendor says
The maker describes it on its own site or launch post. It shows what the product is meant to do, with no independent result.
Real cases of AI agents at work, from every major agent, each linked to where it was first reported. Rewritten in our own words, most with a prompt you can try. These are not Wize customers.