Introducing Agent Effectiveness. Measure Factory usage across cycle time, work priorities, and shipped artifacts, so you can actually see what your AI spend produces.

See where teams are shipping faster. 'Throughput' puts Factory usage next to project, issue, and pull request cycle times. Compare projects, pods, and users against quarterly targets, then pinpoint where the gains are concentrated.

Make sure added capacity reaches priority work. 'Output' maps Factory spend to session intent: feature engineering, maintenance, bug fixes, and exploration.

Trace Factory spend to shipped work. 'Attribution' links sessions to projects, issues, users, and artifacts. See the specific work behind organization-level spend.

Agent Effectiveness is available now in private preview in Factory Analytics:

ROI is back

We are so back!

Love this!

Thanks Shaun!

SO valuable for teams to see how Factory is converting code into valuable, great job team!!!

🫡

Smart move, love this drop 💪

Thank you Josh!

You’re welcome! 💪🔥🫶

Awesome.

Thank you!

Factory Droid Max ($200/mo) feels closer to Claude Max 5x in actual usage. Droid Computer is only ~10h. Could Factory increase the Max quota or let users convert Computer hours into regular Droid quota?

Measuring cycle times, intent, and shipped artifacts provides engineering leaders with tangible ROI metrics for AI spending.

cycle time hides review load, where weak agent output moves cost instead of removing it

Your mobile webpage is very buggy please fix it. @FactoryAI

You need to use cost tracker in the Campfire

measuring actual shipped code against api spend is where most ai deployments stall.

This is actually a big deal. Companies spend millions on AI. But how do they know it’s actually helping? Now they can start measuring what AI actually gets done. And once you can measure the results, AI stops being an experiment and starts becoming a real business tool.

@acgfbr