Occupancy · 05
Forward occupancy
Today is a measurement. Ninety days out is a forecast, and a forecast is only useful if every component is labelled by how it was produced. Three of the five below are measured. Two are modelled, and the model for each is stated.
Today
85.0%
Measured from your inventory
30 day
85.4%
Mostly contracted, low variance
90 day, scenario
72.5%
Modelled components scaled 1.0x
Confidence band
±140 bps
At 90 days, on the model as specified
Ramping centres
8
Under 24 months old, still filling
Net committed change
+70
218 committed in vs 148 on notice, this window
Today to ninety days, decomposed
Slide to stress-test renewal risk and expansion signal
Green adds occupied beds. Red removes them or adds to the denominator. New capacity is dilutive by construction and is the largest single negative in the set.
Magnitude of each component
Absolute bed movement, not direction
5
Steps
Waterfall components
Each step, its size, and how it was produced
| Component | Change | Occupancy after step | Basis | Note |
|---|---|---|---|---|
| Committed move-ins | +4,180 | 201.3% | Measured | Signed and dated. Contracts already in the system. |
| Notices served | -2,470 | 132.6% | Measured | Patients who have given written notice. |
| Renewal at risk | -1,640 | 87.0% | Modelled | Modelled from verified intent captured on renewal calls. |
| Expansion signal | +2,310 | 151.2% | Modelled | Modelled from headcount and bed-need signals inside the patient book. |
| New capacity opening | -3,900 | 72.5% | Measured | Beds entering the denominator before they fill. Dilutive by construction. |
Every centre, forward inputs
Committed move-ins and notices, the two measured legs of the forecast
| Cadabam's Rajajinagar | Bengaluru | 190 | 85.9% | 24 | 15 | 9 | |
| Cadabam's Vidyanagar | Hubballi | 150 | 87.5% | 48 | 15 | 3 | |
| Cadabam's Sarjapura | Bengaluru | 160 | 89.0% | 70 | 14 | 6 | |
| Cadabam’s Bejai | Mangaluru | 160 | 87.7% | 24 | 14 | 6 | |
| Cadabam’s Kuvempunagar | Mysuru | 205 | 84.9% | 71 | 13 | 6 | |
| Cadabam’s Gokul Road | Hubballi | 160 | 83.5% | 42 | 13 | 4 | |
| Cadabam’s Electronic City | Bengaluru | 185 | 90.3% | 68 | 11 | 11 | |
| Cadabam's Vijayanagar | Mysuru | 175 | 83.7% | 14 | 10 | 9 | |
| Cadabam's Kanakapura Road | Bengaluru | 200 | 87.7% | 63 | 9 | 7 | |
| Cadabam's Indiranagar | Bengaluru | 105 | 85.1% | 22 | 9 | 3 | |
| Cadabam’s Banjara Hills | Hyderabad | 125 | 79.7% | 23 | 9 | 3 | |
| Cadabam's Anna Nagar | Chennai | 225 | 85.5% | 66 | 9 | 10 |
What the forecast is telling us
Reading the waterfall and the ramp cohort together
Committed move-ins outweigh notices this window
Measured218 beds are committed to move in against 148 beds on notice — a net positive 70 beds before either modelled leg is applied.
8 centres are still in their ramp window
RampUnder 24 months old and filling toward target — these carry the widest variance in the 90-day scenario because their fill curve is thinner data than a mature centre's.
New capacity is the largest single drag on the ratio
MethodIt adds beds to the denominator before it adds occupants, which is arithmetic rather than a demand signal — worth separating from the renewal-risk conversation on any board pack.
The honest caveat
Neither modelled component existed before the voice layer
Committed move-ins, notices served and the new capacity schedule are already in your systems, and a spreadsheet can forecast from them. Renewal risk and expansion signal are not in any system, because nobody has the hours to ask every account every month what is actually happening. That is the whole argument for the layer, and it is also the reason the ninety day band is wide until the model has been run against your own outcomes for two quarters. The band narrows with data. It does not narrow with confidence.
Demo dataFigures on this screen are illustrative, from a deterministic model, for walkthrough purposes only.