Factory Analytics lets engineering teams examine AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → spending alongside activity and code output. It records model tokenThe chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.Full definition → consumption, tool calls, files changed, commits and pull requests, with breakdowns by user and time period. Its API can also be combined with project-tracker data such as Jira or Linear to relate usage to completed work. Factory describes it as available to Enterprise plan customers. That gives teams evidence for investigating where an agent is used and what accompanies the spending. Interpreting the result still requires care: more edits or pull requests measure activity, and a rise alongside adoption does not by itself establish how much engineering time the agent saved. Analytics announcement
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
