Automating Data Entry for Lyft Commodity

Lyft Commodity: $665K annual benefit by automating manual data entry - hero image

At a glance

Lyft Commodity, a 20-trader commodity trading firm, automated voice-note transcription and trade-system data entry via an AI pipeline built by Rex Automaton. The system eliminated $346,253 in direct annual labor cost, reclaimed 2,400 trader hours per year for revenue-generating work at $133/hr, and delivered a total $665,381 annual benefit at a 192% first-year ROI.

Lyft Commodity's 20 traders generate roughly $250,000/year each, but every one of them spent 30-60 minutes per day transcribing voice notes and meeting recordings into trade systems. At an effective hourly rate of ~$133, that manual entry was costing the firm $346,253 a year in direct labor before counting the opportunity cost of trader time not spent prospecting and negotiating.

We built an AI transcription and summary pipeline that ingests voice notes and meeting recordings, auto-summarizes key decisions, action items, and trade details, tags and routes summaries to the right CRM and dashboard fields, and integrates directly with their downstream trade system. Traders speak. The system writes.