Two things my first look was getting wrong
Condition now comes from the sale record and the photos instead of a data field that called every house heavy. Every saved analysis carries the flood zone, and a failed lookup says so.
THE WORK, BETWEEN ESSAYS
A short note. A changed assumption. Something worth remembering.
Condition now comes from the sale record and the photos instead of a data field that called every house heavy. Every saved analysis carries the flood zone, and a failed lookup says so.
My leads moved onto my phone, texting counts as a contact try, and a lead now finds its own deal. Starting a new deal shows open leads ready to underwrite.
What a seller tells me now gets captured in one place, confirmed field by field, and turned into a plan for the next conversation. It never changes a number.
A flat wholetail percentage was off by about 20 points either way, so each estimate now carries its own wholetail number. Finished jobs get a short debrief on what ran over.
Every bot task carries a real date now, and a filed contract puts its deadlines on my list as open tasks.
The underwriting engine stopped picking a winner. It measures every exit against my buy box and flags thin comps, and the exit choice is mine and on the record.
My content and social bots draft from what happened in the business. Nothing goes out until I say yes, and the newsletter does the sending.
Specialist bots and a thin Chief of Staff now run on Grok with lane skills and Control MCP emit.
Edge holds the shared operating record, with Cowork and the apps working through Control.
Underwriting moved into conversation. The app is back as one screen to review the numbers and make the decision.
Build still handles the scope in the app. Wiring the estimator to conversation is in progress.
Blogs and social content now move through Claude Cowork and publishing skills. I still review and approve what goes out.