Daycare Staffing Ratios: Plan the Daily Schedule With AI

Build the rota from three inputs: your state's ratio table, a half-hour headcount of children by age group, and a grid of staff shifts. Required staff for each slot is the child count divided by the ratio, rounded up. Use a sheet formula for that sum, and use AI to find the slots where scheduled staff fall short and propose fixes.

This is a tutorial about building the rota, not about looking up ratios. Your state's table is an input you paste in and date-stamp. Where this page shows ratios such as 1 adult to 4 infants, they are invented for the example. Never copy them; copy yours from the licensing agency's current page.

Follow me on Instagram@sagnikteaches

What goes into the sheet, and what AI must never supply

Four inputs sit in separate tabs, each with a source and a date. Only one of them needs AI at all, and it is the last.

Connect on LinkedInSagnik Bhattacharya

The reasons for keeping the ratio tab human are the same as for any rule: a wrong figure produces a wrong rota, and a wrong rota is a licensing breach the moment an inspector counts heads. If you have the rules text, you can ask an assistant to pull the ratio table out of it as a way of saving typing, then compare it line by line. The method is the one described in the inspection checklist tutorial.

Subscribe on YouTube@sagnikteaches

Get the headcount by half hour

A rota built on the number of registered children is nearly always overstaffed in the middle and understaffed at the edges. What matters is how many children are physically present each half hour, and that is knowable from the sign-in data you already hold.

1. Export four weeks of sign-in and sign-out times for each weekday, with the child's room and age band but not the name. Use a code or just a row number.
2. For each half-hour slot, count how many children were present, per room.
3. Take the highest of the four counts as the planning figure for that weekday and slot, and keep the average beside it so you can see how often you plan above the typical day.
4. Add known changes: a new enrolment starting, a child moving up a room, a holiday week.

This is a task AI does quickly if the export is coded and uploaded as a CSV: ask it for a table of maximum and average children by weekday, room and half hour. Check two or three cells against the raw data. Spreadsheet pivot tables do the same job with no AI at all, and are fine where the team is comfortable with them.

The formula that decides every slot

In the headcount tab, next to each count, put the required staff. If the child count is in column C and the ratio (children per adult) for that room is in column D:

=IF(C2=0,0,ROUNDUP(C2/D2,0))

If your licensing rules say the building must never have fewer than two adults when any child is present, wrap it: =IF(C2=0,0,MAX(2,ROUNDUP(C2/D2,0))). In the grid tab, count the staff placed in each room per slot with COUNTIF, subtract the requirement, and colour any negative result red. Now a glance tells you whether the day is safe.

Two features of the maths show up in practice. Rounding up means the twenty-first child in a room with a 1:10 ratio triggers a third adult. And the count must be per room unless your rules allow adults to be pooled across rooms in the same slot, so the grid should assign each person to a room, not to "the floor".

Main worked example: one Tuesday in a three-room daycare

An illustrative daycare opens 7:00 to 18:00 with three rooms. The invented ratios are 1:4 for infants, 1:6 for toddlers and 1:10 for preschool, and the rule for the combined early and late rooms is that the youngest child present sets the ratio. Planning headcounts from the sign-in data:

Notice that the preschool room is exactly at the limit at peak (20 children, two adults). Any extra child in that window needs a third adult, which is worth a note on the grid. The director then proposes eight shifts of 8.5 hours with a 30-minute unpaid break, and asks the assistant to check them:

Shifts (all with one 30-minute unpaid break):
S1, S2: 07:00-15:30   S3, S4: 08:00-16:30
S5, S6: 08:30-17:00   S7: 09:00-17:30   S8: 10:00-18:30
Required adults by window: [paste the table above]
Task: for each half-hour from 07:00 to 18:30, count staff on shift
(ignore breaks for now), compare with required adults and list every
slot where staff on shift is lower. Show your counting for each slot.

The answer, once checked, flags two slots, both in the last hour of opening: from 17:00 to 17:30 only S7 and S8 are on shift, two adults against a requirement of three, and from 17:30 to 18:00 only S8 is left, one adult against two. Everything earlier is covered: two on shift from 07:00, four from 08:00, six from 08:30, seven from 09:00, eight from 10:00 to 15:30, then six until 16:30 and four until 17:00. That is the kind of finding that is easy to miss by eye, because the morning and the middle of the day look comfortable and the trouble sits at closing time.

A realistic slip to watch for: a first answer that reports only the 17:00 slot has stopped counting when S7 goes home at 17:30. That is why the prompt asks the tool to show its counting for every slot. Read down the column yourself for the final hour; it takes a minute, and closing time is where a gap like this is easiest to miss.

The fix is a small one, and it closes both slots. Add a part-time shift S9 from 17:00 to 18:30, one and a half hours, which gives three adults from 17:00 to 17:30 and two (S8 and S9) from 17:30, matching the requirement of two. At an illustrative $18 an hour, that costs $27 a day. The alternative, moving one of the 08:30 starters to 10:00, would cover the close but open a gap at 08:30, five adults against six, so the AI is asked to test the alternative too and report that it fails. The point of the assistant here is speed and patience at checking, while the choice is the director's.

Breaks are where ratio is lost

The grid above ignored breaks. Reality does not. Someone on a break is not supervising, so they leave the count for that slot. With seven on shift from 09:00 and six required, there is exactly one spare adult, which means one person on a break at a time. After 10:00, with S8 on, there are two spare, so breaks have more room.

A workable break plan for the example runs one at a time from 11:00: S1 at 11:00, S2 at 11:30, S3 at 12:00, S4 at 12:30, S5 at 13:00, S6 at 13:30, S7 at 14:00 and S8 at 15:00. The break window is the middle of the day because that is when there is a spare adult. Two rules help. No two breaks overlap in the same room. And the person on break is replaced by the spare, who moves to the room that needs cover, which is a room the grid should name.

Ask the assistant to place the breaks as a second pass on the checked grid and to list any slot where the number of adults in a room, with the break removed, falls below the requirement. Where your rules treat nap time differently, for example allowing a different ratio while children sleep, put that rule in the ratio tab with its source, and never let the assistant assume it.

A cover list for the morning someone rings in sick

Absence is the day-to-day risk. A cover list takes five minutes to build and saves the worst phone calls of the week. Columns: code, qualification level, which rooms they can cover, days available, hours they would accept, how soon they can arrive, and an order to call. Keep the phone numbers in your own system, not in the tool.

When the call comes in at 6:40 that S3 is ill, paste the day's grid and the coded cover list, and ask:

S3 is absent today. Using the grid and the cover list, give me the
smallest set of changes that keeps every room at or above the required
adults in every half-hour slot. For each option list the staff moved,
the slots affected, and anything that needs a decision from me
(overtime, a shorter break, a call to someone off shift).
Do not assume anyone is available who is not on the cover list.

Illustrative output: Option A, call C2, available 08:00 to 16:30 for preschool, who covers S3's shift exactly. Option B, if C2 is not reachable: without S3 the day falls one adult short from 08:30 to 09:00 and from 15:30 to 16:30, so ask S7 to start at 08:30 (half an hour of overtime) and move the part-time S9 to start at 15:30 instead of 17:00. That holds every slot, but leaves only one spare adult between 10:00 and 15:30, so breaks must run strictly one at a time, and S9's longer shift needs your approval. The useful thing is the second option with its costs stated plainly. You make the calls.

A smaller case: twenty children and the two-adult rule

Not every daycare has three rooms and nine staff. Take an illustrative setting with a single room of up to 20 mixed-age children, open 7:30 to 17:30, and an invented ratio of 1:6 for the youngest age present. Suppose the licensing rules, or your own safeguarding policy, say that at least two adults must be present whenever any child is on the premises. Then the arithmetic changes shape. For up to 12 children, ROUNDUP(12/6) is 2, so two adults cover it. At 13 children the figure becomes 3. The first and last half hours, with perhaps three children, still need two adults, even though the ratio alone would allow one.

That minimum is what makes the opening and closing hours expensive in a small setting: two people paid to look after three children. The sheet shows it plainly. Suppose the grid shows two adults from 7:30 to 8:15 with an average of 2.4 children. The director can ask whether the first slot could open at 8:00, after asking parents whether anyone relies on it. That is a business decision the sheet informs; it does not make it. If you lack a second adult at the edges, the sheet is also where you notice a single point of failure before it becomes an incident.

A tool can help here with a "what if" in one prompt: "If we close at 17:15 instead of 17:30, how many child-minutes of care and how many adult-hours would we lose on a typical Thursday?" Give it the headcount tab and let it do the sums, then verify with a quick formula before showing anyone.

Shift patterns that fit a ratio curve

Children arrive in a hump, with a flat middle and a tail, while staff shifts are blocks. The art of a ratio-compliant rota is stacking blocks so they sit just above the curve. Three patterns work for most daycares:

• Staggered starts. Two staff open, two more join at 8:00 and 8:30, as the building fills. This matches the morning ramp without paying for four adults at 7:00.
• The late floater. One person starts at 10:00 and finishes at 18:30, covering the middle breaks and the tail. In the worked example this is S8, and the short S9 shift handles the last ninety minutes.
• Split-day part-timers. Two short shifts for staff who want 6 to 7 hours can cover the lunchtime break window without lengthening anyone else's day.

Give the assistant the headcount curve and the shift menu and ask for three candidate rotas that never drop below the requirement, with total paid hours for each. Check the best one in the sheet. Ask the assistant to explain why the other two lose: the answer tends to reveal an assumption, such as a break rule you forgot to mention, that is worth correcting before you commit.

Five ways an AI ratio check goes wrong

• The wrong ratio for a mixed room. A combined room set by the youngest child present is the usual rule in examples, but your rules may weight differently. Test the assistant with a case you can compute by hand.
• Counting people who are not in ratio. The cook, the director while doing admin, a volunteer or a student may not count. Add a "counts in ratio" column and make the tool use it.
• Double counting overlapping shifts. Chat tools sometimes count a person twice where two shift rows overlap. The COUNTIF formula in the sheet does not, which is one reason the formula decides.
• Rounding down. A tool may say 9 children at 1:4 needs 2.25 adults and call that "about 2". The rule is 3. Say "round up" in every prompt.
• The outdoor and outing blind spot. Ratios on trips, in the garden or at the park may differ from inside. If your rules have separate figures, give them their own rows.

Check the headcount at the peaks, not on average

A rota can pass in the grid and fail at the door because the real peaks are sharper than the planning figures: a wave of arrivals at 8:45, a clump of early collections at 15:30 on Fridays. Once a week, take the sign-in export and test it against the rota: for every slot where the real count was higher than planned, work out whether the staff on shift still met the requirement.

A quick audit, as an illustration: in a month, 160 slots were checked and 3 slots had a real count that would have needed one more adult than the plan. Each was a Friday at 15:30. The remedy was a shift move of 30 minutes on Fridays and a note on the grid that Friday departures bunch up. Ratio compliance on actual data, not on plan, is the figure an inspector cares about. The related nursery rota tutorial shows a similar weekly audit in a nursery setting, and the small-team scheduling comparison helps you decide whether a spreadsheet or a scheduling app suits you.

What you tell staff, and what you keep private

The grid becomes a staff-facing rota. Ask the assistant to produce a plain weekly view: names replaced from your code list by you, rooms, start and finish, break times. Draft the message that goes with it, including how to swap shifts (both people agree, the director approves, the ratio check re-runs). Keep availability, qualification levels and any personal reasons for leave out of the shared copy.

On privacy generally: headcounts and staff codes are low-risk, but real names, children's names and pay details are not material for a chat tool. Staff codes in, real names added afterwards in your own sheet.

What it costs you and what it saves

For a setting like the illustrative one above, the first build takes a morning and weekly upkeep about 20 minutes: update the headcount from last week's data, paste the absence list, re-check any changed slots. The benefit is that you find breaches on a Tuesday afternoon with a sheet instead of on an inspection morning with a clipboard. A secondary benefit is cost. A grid that shows surplus adults in the quiet window from 10:00 to 15:00 is also a map of where hours can be trimmed or moved to the edges where you are short. Move two hours a day from surplus to shortage and you have fixed a risk at no extra payroll.

If you would like the sheet built around your own rooms and software, that is what a 1:1 AI implementation consultation would cover.

The decisions that stay with the director

Which ratio applies in an unusual case such as a child with additional needs, when to exceed the minimum for safety rather than compliance, and whether an overstretched day should have a room closed. The sheet will say what the rule needs. Only a person standing in the room can say what the children need.

Ratio and rota questions

Can AI tell me my state's ratio?

Not reliably from memory. Ratios depend on the state, the setting type and the age of the children, and they change. Copy the table from your licensing agency's current page, paste it into your own sheet, and let AI work from that text only. Check the first few results by hand.

Why do the arithmetic in a spreadsheet and not in the chat?

Chat tools sometimes slip on sums, rounding and counting overlapping shifts. A formula never does. Use the AI to design the grid, explain gaps and propose fixes, and use ROUNDUP formulas to produce the required-staff numbers that decide whether a slot is safe.

Do breaks count as staff in the room?

No. Someone on a break is not supervising, so remove them from the count for that slot. Plan break cover from the surplus at quieter times, and never let two breaks overlap in a room that is already at the minimum.

How do I handle a sudden absence?

Keep an ordered cover list with each person's availability and qualifications, and decide at what time you start calling. Paste the day's grid and the list into your assistant and ask for the smallest change that restores ratio in every slot. Check it, then phone people yourself.

Further reads

• Choosing Nursery Software With AI: What to Compare — If a spreadsheet is not enough, what to compare in software.
• Retail Staff Scheduling: Build a Rota From Hourly Till Sales With AI — Building a rota from an hourly demand curve, in retail.
• How to Build a Restaurant Staff Rota With AI Demand Forecasts — How demand forecasts shape a rota, in another industry.
• Daycare Late Pickup Policy: Can AI Word the Fees Fairly? — Late pickups stretch your closing cover; set the policy.
• AI for Daycares: Nine Admin Jobs From Enquiry to Pickup Message — More daycare admin jobs worth handing to AI.
• Artificial Intelligence Workforce Automation: Rotas, Timesheets, Leave — Link the rota to timesheets and leave requests.
• Brightwheel vs Procare: Which Daycare Software Has Real AI? — Brightwheel's AI assistant against Procare's RoomRunner and Camber, with quote-only pricing questions and a scored demo for a small daycare.
• Urgent Care Staffing: Forecast Volume and Build Rotas With AI — Turn a year of visit counts into an hourly volume forecast and a draft rota, with the manager's checks that keep it safe.
• Child Care Subsidy Billing: Can AI Match Payments to Attendance? — How a daycare can match subsidy remittances to attendance with AI-built formulas, find shortfalls and write top-up notices without names.
• How to Start a Preschool: First-Year Costs and Where AI Fits — A first-year cost table for a small preschool, the licensing route, enrolment targets, and the forms and parent messages AI can draft from day one.
• AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.

Sources: your own state's child care licensing rules and ratio table (check the current page for your setting type and age bands); Google Sheets function list for ROUNDUP and COUNTIF (checked November 2026). Ratios used in examples are invented for illustration.