How to Increase Nursery Occupancy: Open Days, Reviews and AI

A nursery raises occupancy by improving four steps it controls: how many enquiries it receives, how fast it replies, how many families visit, and how many visitors are followed up. Open days, quick follow-up and calm review replies each move one step. AI drafts the messages; the manager sends them and checks the counts weekly.

Occupancy is a result, not an action, so chasing it directly rarely works. A nursery with 40 places and 34 children is at 85%. To reach 38 children, it needs four more families to sign. Working backwards through the funnel shows where to spend effort: if 60 enquiries a term produce 12 visits and 4 places, the weak step may be getting enquirers to visit, not getting more enquiries.

Follow me on Instagram@sagnikteaches

Measure the funnel for four weeks before changing anything

Most nurseries know how many children are on the roll and have no idea what happened to the families who did not join. A simple record fixes that. Use a shared spreadsheet, not a person's memory, and put one row per enquiry: date received, source (website, phone, referral, review site, social, open day), date of first reply, whether a visit was booked, whether it was held, and the outcome (place offered, accepted, declined, no reply). Do not record a child's name, health or needs in this sheet; a reference number and the room age band are enough, and the details stay in your registration system.

Connect on LinkedInSagnik Bhattacharya

After four weeks, add up four numbers and then the three ratios that follow from them. Here is an illustrative month for a 40-place nursery:

Subscribe on YouTube@sagnikteaches

In this example, 13 enquirers never booked a visit, 3 booked and did not attend, and half of those who visited did not join. Each loss has a different cause and a different fix, which is why the counts come first. The 46% booking rate points to reply speed and clarity. The 73% attendance rate points to reminders. The 50% join rate points to the visit itself and the follow-up after it. If you can also see how long the first reply took, ask a chatbot to compare the booking rate for enquiries answered within an hour against those answered later. Paste only dates and counts, never names.

Plan an open day around the questions parents bring

An open day gives a family several things at once: a look at the rooms, a feel for the staff, and a chance to ask what they were embarrassed to put in an email. A plan that treats it as a show loses to a plan that treats it as a conversation.

A sensible structure for a two-hour session offers staggered 20-minute slots, so each family has a tour with a manager or room leader and time to talk, rather than a crowd in a corridor. Staggering also means you can show the rooms with children present at the right times without stopping care. Check your safeguarding rules for visitors, and make sure visitors sign in, stay accompanied and are never alone with a child.

Let AI draft the plan, with your constraints

Plan a two-hour open morning for a nursery with 40 places and four rooms (baby, toddler, pre-school, outdoor area).
Constraints:
- Visitors arrive in 20-minute slots, up to 3 families per slot
- A manager or room leader leads each tour; visitors are always accompanied
- No photographs of children by visitors
- We cannot disrupt snack time at 10:15
Give me: a run sheet by time, a checklist of what to prepare the day before, three questions we should ask each family at the end, and a short "what to bring" note for parents.
Do not suggest any claims about outcomes or inspection results.

A reasonable answer gives a run sheet, a tidy checklist and some good questions ("What matters most to you in a first setting?", "When would you need a place from?"). Fix three things: remove any suggestions that conflict with your own safeguarding rules, change times to fit your daily routine, and check that the "what to bring" note says nothing about what you cannot provide.

The invitation and the reminder

People forget open days. Draft two messages: an invitation sent a fortnight ahead, and a reminder the day before with the slot, the entrance, parking and what to expect. Add a sentence that makes it easy to cancel: "If you cannot make your slot, reply and we will offer another." Families who can cancel kindly are more likely to rebook than those who feel embarrassed and vanish. A visit-booking assistant can take over the simplest of these steps; the tutorial on AI chatbots for nursery enquiries and visit bookings covers what to put in it and what to keep for a human.

Follow up every visit the same day

This is the step with the biggest swing in the illustrative table above, because half of visiting families did not join. A family that has seen the nursery is warm, but their memory fades quickly and they may visit three more settings that week. A same-day message does three jobs: it thanks them, answers the questions they asked on the tour, and makes the next step easy.

The way to make same-day follow-up realistic is a template, not a blank page. During the tour, the manager notes two or three things the parent said (a preference for outdoor play, a worry about settling, a start date). That evening, those notes go into a prompt with no names or identifying details.

Write a follow-up email (under 130 words) to a parent after a nursery visit.
Notes from the tour: they liked the outdoor area and the key-person system; they worried about settling a shy child; they want a start in September, three days a week.
Include: thanks, a direct reply to the settling worry (our settling sessions are gradual and agreed with the parent; no specific promises), what to do next (reply to say if they would like a place, or call [phone]), and the date we will hold their enquiry until: [date].
Do not mention fees or funded hours; say we will send a price and funding guide separately.
Use [first name] and [your name].

A usable draft is warm and specific. Three things need a human check. Make sure it does not promise a place, because availability and registration rules decide that. Check the "hold until" date is real. Remove any detail about the child you would rather not have in an email thread. Then send it before the end of the day, and put a reminder in the diary for three days later.

The second nudge

If the parent has not replied in three days, send a short second message: "Hi [first name], I just wanted to check whether you had any questions after the visit. We still have availability on the days you mentioned, and I am happy to talk it through by phone. [your name]". That is helpful, not pushy. After that, record the outcome and stop. A family that chooses another nursery has made a decision, and a polite goodbye leaves a good impression.

Check funded-hours wording before it goes out

Many families look for funded hours or funded places, and a nursery's wording about them has to match the scheme's rules. A careless sentence such as "all our places are free" or "your child's hours will all be funded" can mislead and cause disputes later. AI will write smooth sentences about funding that sound right and are wrong for your setting, so treat funding text as something a person with the facts must approve.

Use this routine. Take the current guidance from the body that funds your places, paste the exact paragraphs you rely on into the chatbot, and ask it to draft a plain-English paragraph for parents, using only that text. Then compare the result line by line with the guidance, and ask whether the parent needs to take any step (such as applying or confirming eligibility). State plainly what the family still pays for: meals, consumables, additional sessions, or any charges your funding agreement allows. If you are unsure, ask the funder in writing and keep the answer. The tutorial on fees, deposits and late payment wording carries the same care into your policies.

Reply to reviews in a way that reassures the next parent

A review reply is written for two readers: the reviewer, and the dozens of parents who will read the exchange before they book. A calm, specific reply to a positive review shows the nursery is attentive. A calm reply to a negative one shows maturity. An argumentative reply does damage that no open day repairs.

Set a simple rule: a manager reads and sends every reply, within two working days. AI helps with the first draft, and especially with tone. Before pasting a review into a chatbot, check your privacy rules: the review is public, but the reply must never confirm that a named child attends or reveal anything about them. The tutorial on whether to let AI auto-post review replies explains why auto-posting is a poor idea.

Draft a reply (under 90 words) to this public review of our nursery:
"We were disappointed that nobody got back to us for four days after we asked about a place."
Rules: thank them, apologise for the delay in a way that is sincere and specific, do not say whether they or their child attend, do not give any detail about a child, invite them to ring [phone] to talk, do not argue or explain at length.
Tone: calm, human, not corporate.

A typical output apologises, says "we aim to reply within a working day" and invites a call. Check that last sentence: only promise what you can deliver. If the review is unfair, the reply can say so politely, once, without detail. If the review raises a safeguarding concern, do not reply publicly in the draft stage; follow your safeguarding procedure first and take advice. Positive reviews deserve a reply too, a short thank-you that mentions something specific and general ("thank you, we are glad the settling sessions helped") but never a child's name.

Asking for reviews the right way

Ask parents who have been with you for a term, using a short message and a direct link. Do not offer discounts or gifts for a review, and do not draft reviews for parents to copy. Google's contribution policy, for example, bars incentives for any review and bars selectively soliciting positive reviews or discouraging negative ones, so ask every family at the same point rather than only the ones you expect to be pleased; check the policy of any other platform you use. The end of a first term is a natural moment, and a note on a pick-up day works as well as an email.

Make the first phone call count

Many parents ring before they email, and the person who answers shapes the whole funnel. A manager who is on the floor with children cannot always pick up, so decide in advance what happens to calls you miss. A clear voicemail message that names a call-back time, plus an instant text reply with a link to book a visit, rescues many of them. If your phone system transcribes messages, paste the transcript (with the caller's name and number removed) into a chatbot and ask for a two-line summary and a suggested call-back opening. That turns a pile of voicemails into a short list the first person in on Monday can work through before 9.30.

Give whoever answers a one-page fact sheet: age ranges, opening days and hours, how a visit works, what the settling sessions are, and what to say about fees and funding ("we will send our current price and funding guide after you visit"). Ask AI to tidy the sheet, then have the manager approve every line. When the facts change, change the sheet first, so nobody quotes last year's hours.

Fix the three website pages parents read first

Before a family books a visit, it usually looks at the website. Three pages matter most: the home page, the page about the rooms and ages, and the page about visits and fees. For each, ask whether a parent could answer these questions in under a minute: Do you take my child's age? Which days and hours? How do I book a visit? What happens when I get there? A chatbot is useful as a first reader here. Paste the page text and ask it to list the questions a worried parent would still have after reading it, then answer the top five in a short question-and-answer block on the page.

Take care with claims. Inspection outcomes, ratios, qualifications and funding statements must be current and match your records, and the wording must stay within your regulator's advertising rules. If a sentence feels too good to be true, check it against the document it came from. If your pages mention photographs of children, make sure you have written consent from the parents and use the minimum needed.

A before and after for a visit-booking page

Before: a page that says "Contact us to arrange a viewing" and a phone number that rings in the office. After: a short page with three visit slots on two days, a form asking only for a parent's name, contact details and the child's age band, and a line saying "we reply within one working day". Illustrative effect: a nursery that moves from 46% to 60% visit-booking rates on the same 24 enquiries books about three more visits a month. That is a guess to test, not a promise, and the weekly counts will show whether it holds.

A weekly review that takes 20 minutes

Put a standing slot in the manager's diary, say Friday at 2pm. Update the sheet with the week's enquiries and outcomes, then paste the counts (no names) into a chatbot and ask for three things: the funnel ratios, which step moved since last week, and one change to try next. Treat its answer as a prompt for discussion, not an order. A week with two enquiries tells you nothing, and the chatbot should say so; if it does not, tell it to.

Here are this week's counts for our nursery (no names): enquiries 7, first reply within one hour 5, visits booked 3, visits held 2, places accepted 1. Last week: enquiries 5, reply within an hour 3, visits booked 1, visits held 1, places accepted 0.
Give the ratios for each week, say which step improved, say what we cannot conclude from such small numbers, and suggest one small thing to test next week.

A good reply notes that the numbers are small, that two weeks cannot show a trend, and that the one change worth testing is the reply time. It may suggest comparing four-week blocks instead. Take that advice. Each month, compare the block with the previous one and write one sentence about what you learnt.

Use local groups and your own families

Parents trust other parents, and most nurseries have families who would speak well of them. Local parent groups, toddler classes, library story sessions and community family centres are natural places to be known. Ask permission before you leave posters or take part, share news rather than adverts (a new outdoor garden, a taster session, a date for the next open morning), and respect each group's rules on promotion. Draft the post with AI, then read it as a stranger would. If it sounds like an advert, shorten it.

A simple "invite a friend" note to current families can work, as long as it does not offer rewards that breach your funding terms. A line such as "if you know a family looking for a place, we would be glad to show them round" costs nothing. For posters, flyers and open-day graphics, the tutorial on Canva plans for a tutor or nursery says which plan is worth paying for.

What changes in a month, with illustrative numbers

Return to the 40-place nursery at 34 children. Say it makes three changes: replies to enquiries within an hour on weekdays, a same-day visit follow-up with a second nudge, and a monthly open morning with staggered slots. After a month of the new routine, the illustrative table might read: 26 enquiries, 16 visits booked (62%), 13 visits held (81%), and 7 places accepted (54%). That is three more families than before, which is a 7.5 percentage-point rise in occupancy once they start. Real results vary, and a month is too short to prove anything, so keep the weekly counts running for a full term before you claim a cause.

Notice what AI did and did not do in that picture. It drafted the open-day plan, the invitation, the follow-up and the review replies. It did not decide who gets a place, set a fee, or talk to a parent. Those jobs stay with people, and they are the reason a family trusts the nursery.

Where measured follow-up goes wrong

• Speed over accuracy. A fast reply that quotes the wrong fee or hour is worse than a slightly slower correct one. Keep a fact sheet your team and your AI prompts both use.
• Children's details in prompts. A tour note such as "shy, speech delay" is sensitive. Describe needs generally, or leave them out of anything you paste, and follow your own data rules. The tutorial on using AI with children's data safely covers the line.
• Counting only enquiries. A busy month of enquiries that never visit can hide a broken booking step. Count all four numbers.
• Marketing promises you cannot keep. "Top inspection rating" or "limited places left" must be true on the day it is read. Check the claim, the date and the source before publishing.
• A review reply written in anger. Draft it, leave it an hour, and let a second person read it.

If you would like help setting up the weekly counts and the follow-up drafts around your own software, that is what my AI implementation consultation covers. The tutorial on choosing nursery software with AI is a good companion if your records live in several places.

Questions nursery owners ask about filling places

How fast should a nursery reply to an enquiry?

As fast as your team can do it properly. Aim for an acknowledgement within an hour in working time and a full reply, with visit slots, the same day. Parents often contact several settings at once, so a prompt, clear reply is one of the few parts of the funnel fully in your control.

Can AI answer parents' enquiries on its own?

For simple questions such as opening hours and how to book a visit, a chatbot or autoreply can help. Anything about fees, funding, a child's needs or availability should be checked by a manager. Never put a child's name or details into a public chatbot.

Should we reply to negative reviews?

Usually yes, briefly and calmly, within a couple of days. Thank the reviewer, avoid confirming whether a child attends, offer to talk privately and do not argue. A manager should read and send every reply, whether or not AI drafted it.

What counts should we track each week?

Four are enough: new enquiries, visits booked, visits held and places offered or accepted. Add the source of each enquiry if you can. Weekly counts show whether a change helped, and they stop one busy fortnight from steering every decision.

Further reads

• Tapestry vs Famly for a Small Nursery: Where Does AI Fit? — Nursery software and where AI fits in it.
• Can AI Write a Nursery Self-Evaluation Form From Your Evidence? — Use the same evidence to support your self-evaluation.
• Nursery Late Collection Policy: Wording, Fees and Parent Messages With AI — Policy wording for a common parent frustration.
• Nursery Staff Training: A One-Hour AI Briefing on Children's Data — Brief staff before they paste anything into AI.
• Early Years AI Policy Template for Nurseries and Childminders — An AI policy to adopt before you start.
• Nursery Data Protection Policy: Draft and Update It With AI — Keep your data policy current as enquiry volumes grow.
• Franchise Review Management: Replies for Several Stores, On Brand — A store-level workflow for franchisees: a reply library, escalation rules and per-site fact cards so AI drafts stay on brand and a person approves each reply.
• Nursery Welcome Pack: Letters and Settling-In Guide With AI — What goes in a nursery welcome pack, with prompts for plain-English letters, a settling-in guide, a key person introduction, forms and checked translations.
• Preschool Enrolment Form: Build, Send and Chase Forms With AI — Draft the fields, consent wording and reminder emails with AI, then chase missing documents from a simple tracker that holds no child details in the chat.
• Preschool Newsletter: A Monthly Issue From Your Notes, With AI — A staff notes template, one reusable prompt and a pre-send checklist turn scattered monthly notes into the same dependable family newsletter every time.
• Preschool Social Media: Safe Posts and Photo Consent With AI — A consent register, a short rules list, caption prompts and a two-person approval step let AI plan a month of posts without exposing a single child.
• AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.

Sources: the nursery's own enquiry and occupancy records, the funding body's published guidance on funded hours (reader to check current wording), and Google's contribution policy for reviews (checked November 2026).