Work-order enrichment: what the robot pulls and how it should scale

A field-by-field map of the AppFolio work-order enrichment robot, how it runs today, and three options for scaling it past Camellia — written for Gera.

2026-09-12 · Fable for Gera · handoff, proposed

The verdict, in plain English

The robot pulls real fields Clara needs before dispatching a vendor — access permission, entry instructions, preferred time windows — that AppFolio's cheap report feed simply does not carry. That part is working and is not urgent.

The gap Fede's worried about is real, but it isn't the robot's fault: nothing today stops Clara from dispatching when BOTH AppFolio and the tenant conversation are silent on access. That's a decision-logic gap, not a data-freshness one.

At 10x the portfolio, the current once-a-day catch-up sweep runs out of runway inside its own time window and staleness would double. The fix: move the fields AppFolio's cheap report already carries (vendor, schedule) off the robot entirely, and re-enrich the rest only when something actually changes — not on a fixed clock.

Nothing here is live or urgent. This ships dark, proven at Willows, per the standard playbook — turning anything on for a real property stays Fede's call.

PropFlow Docs