Case studies

What we found, and what we built.

Two write-ups. One is a real engagement, where the assessment found the job that was eating the week and named the fix. One is a system I built myself, running against a fictitious company, with the log of everything it got wrong still attached.

What we found

Sometimes the answer isn't AI

Troy S. is a manufacturer's rep for specialty building products. Every substitution request meant 30 to 60 minutes assembling the same package by hand, so most simply didn't get done. He now reviews each one in about five minutes. Roughly 7.5 hours back every week, and none of it was AI.

Read what the assessment found

What we built

The agent that works the aging report

Reference implementation · fictitious company

It reads down an aging queue, drafts a follow-up in the tone each customer has earned, sends the risky ones to a person with the briefing already written, and holds every single email for approval. It runs against a fictitious company I invented, not a client's books. The write-up includes what it got wrong and every assumption behind the numbers.

Read how it was built

Start

Your week has 5 hours hiding in it.

Both of these started the same way. One call, one report, one walkthrough. $999, refunded in full if the report doesn't show you a path to 5 or more hours a week.

Start with the Assessment · $999

Report in 3 business days, walkthrough the day after.