Case studies
What happened when someone called us in, and what we told them.
Weeks of failed automation, fixed in two and a half hours
An engineering college at a major public research university spent weeks trying to make an AI flow generator build their entire budget approval workflow. The fix took an afternoon and needed no new software.
Read the case studyWe went shopping for a meeting tool. We already owned it.
Our messages were split across Signal and Teams, and we wanted AI transcripts. We trialed Read AI, then moved to Google Chat, Meet, and Gemini, already in our Workspace, and didn't renew Teams.
Read the case studyHundreds of bills and receipts, one naming rule, and no LLM
One of us renamed about ten years of bills and receipts (utilities, subscriptions, loans, and the mortgage) into one format. An LLM was too costly, slow, risky, and inexact, so we built rules and OCR instead.
Read the case study