If you operate senior living building(s), every staffing conversation seems to loop:

“Hire more.”
“Cut agency.”
“Hold people accountable.”

None of that answers the real question:

Do you actually have enough coverage for the care you’re delivering,
on this shift,
for these residents,
with this team?

If you look back at a few rough weeks, two patterns usually show up:

• Understaffed hours → the day slides: care gets delayed, call-lights stack, behaviors escalate.

• Overstaffed hours → the day feels fine, but payroll creeps and NOI quietly drops.

Same building. Same headcount. Different hours.

Why the old model keeps failing you

The old model is built around averages:

“One aide per X residents on this floor.”
“Y FTEs for Z census.”

But your hardest days aren’t “average.” They’re peaks.

Here’s what that looks like in real life:

7–10 am on 1st floor:
med pass + breakfast + three heavy assists + one refusal + one family question.

By 9:15, showers are slipping.
By 10:30, you’re “catching up” the rest of the day.

If you staff evenly by floor or by census, you’re almost guaranteed to be wrong at the exact times that matter most.

That’s how you end up “fully staffed” and still behind.

Not because the team isn’t working, but because the plan doesn’t match when the work lands.

What “good staffing” means in 2026

What’s worked better in buildings I’ve worked with is simple:

Staff by acuity peaks, not by census or floor.

Acuity peaks are those 2–3 hour windows for a resident (or cluster of residents) where, if you’re light, the whole day tilts.

That’s where:

• overtime starts

• care gets rushed

• the “we’re drowning” feeling comes from

Staffing “right” in 2026 doesn’t mean a perfect ratio.

It means you:

• know when each high-acuity resident has their peak window

• know what usually goes sideways in that window

• put extra hands into that stretch, instead of spreading them evenly

Where to use this this week?

You don’t need software for this. Give it one week in one building.

1) Make a peak map for your high acuity residents
Identify your top 10 high-acuity residents.

For each resident, write their 2–3 hour peak window (7–10, 10–1, 1–4, 4–7, etc.).

Inside that window, tag what’s driving time in three buckets:

Planned: the known routine work
(ex: meds, toileting rounds, showers)

Unplanned: the repeat “surprises”
(ex: PRNs, behaviors, extra assists, incident follow-ups)

Admin: the work that pulls staff off the floor
(ex: charting, family calls, coordination, hunting supplies, policy lookups)

2) Then staff the overlap, not the floor.
When 4–6 residents have peaks in the same 2–3 hour block, that’s where you add hands.

Not evenly across the floor. Not “because that’s the assignment.”

3) Make one move for 7 days.

• add a floater only during that peak block

• pre-assign a backup for the residents with Heavy Unplanned

• move non-time-sensitive admin out of the peak

Run that for 7 days and look back at your notes.

You’ll start to see:

• the same hours where you’re truly short,

• the hours where you’re clearly over,

• and the “unplanned” work that isn’t random at all; it shows up almost every day and should be built into the plan.

That’s the 2026 shift:

From staffing by census and floor,
to staffing by acuity peaks and real workload.

-Kamal Bhartiya, Founder & CEO at Fitmedik

If you’re running your building on ALIS or PointClickCare, my team’s happy to get your building(s) started with AI, without ripping anything apart.

To see what’s working for other operators and identify your first AI-delegated workflow, book a 15-minute brainstorming session here.