Product · Occupancy OS
Occupancy OS
Eighteen screens fusing occupancy truth, demand intelligence, patient-book risk and a voice layer into one console for Cadabam's — built so that every figure can be defended, not just displayed.
Portfolio beds
3,595
24 centres · 8 cities
Portfolio occupancy
85.0%
Blended, mature + ramping
Value of 100 bps
₹0.4 cr
Annualised, at assumed ARPD
Patient NPS
79
Portfolio-wide
Occupancy, 9 months
85.1%
▲ 7.8%vs Jan
Enquiries, monthly
2,140
▲ 32%vs Jan
Patient NPS
79
▲ 10%9-month climb
Modelled revenue
₹29.4 cr
▲ 22%vs Jan, ₹ cr/mo
What this module is
Read this before the screens
Occupancy OS is a single console spanning five layers: occupancy truth, demand intelligence, the existing patient book, a voice layer that reads every conversation, and a method section that shows exactly how each number is produced and where it could be wrong. It is read-only until switched on, and nothing here pretends to be more certain than it is.
Portfolio occupancy and enquiries, trailing 9 months
Two series, one axis each, indexed to the demo ladder
Beds by city
Where the portfolio actually sits
3,595
Total beds
Where to look first
Agent-ranked, cross-module signals
Occupancy has climbed 7.8 points in nine months
OccupancyMature centres are running further ahead than the blended number shows — see Portfolio position for the mature-only split.
Enquiry volume is up a third, response time is not
DemandThe referrer and enquiry-to-bed screens both point at the same bottleneck: speed of first response, not volume of leads.
Renewal gap flags beds no risk list has caught yet
Patient bookSixteen accounts loaded in this preview show a meaningful split between community rating and voice-verified intent.
Nine boundaries are published, not just promised
MethodThe refusal screen lists exactly what this layer will not do, so it can be checked rather than taken on faith.
Occupancy
What is occupied, where, and where it is headed
Demand
Every enquiry, call and channel, reconciled
Patient book
Expansion, renewal risk, service and yield inside the book you already have
Voice layer
The conversation itself, annotated
Method
How every number is produced, and where it could be wrong