Funding claims are getting more frequent. Your spreadsheet isn't getting any smarter.
From April 2026, the DfE is introducing termly head counts for three- and four-year-olds — increasing the number of funding returns providers need to submit. On top of that, you're tracking 15-hour and 30-hour entitlements, two-year-old funding, tax-free childcare, stretched and unstretched hours, and each local authority has its own portal, deadlines, and data requirements. Most nurseries manage this in spreadsheets and the manager's head. Your funding claims agent tracks attendance against entitlements, calculates claim amounts, flags discrepancies before submission, and generates returns in the format your local authority requires — so claims are accurate the first time, not after a clawback.
What Your Agent Actually Does
Your agent tracks entitlements, calculates claims, and generates accurate funding returns — so you claim what you're owed and nothing comes back as a clawback.
Tracks attendance against funded entitlements
15-hour universal, 30-hour extended, two-year-old funding, stretched hours — your agent maps each child's attendance against their specific entitlement and flags when hours are under- or over-claimed. No more manual cross-referencing between registers and funding spreadsheets.
Generates LA-specific funding returns
Every local authority has its own format, portal, and deadlines. Your agent produces returns in the format your LA requires — headcount data, hours breakdowns, eligibility codes — ready to submit without reformatting.
Handles the new termly head counts
The April 2026 change to termly head counts for 3-4 year olds means more returns, more often. Your agent generates each one automatically from your attendance data — no additional workload for a change that was supposed to make funding more accurate, not more burdensome.
Flags discrepancies before submission
Eligibility code expired? Hours claimed exceed entitlement? Child's attendance doesn't match what was registered? Your agent catches these before you submit — not when the LA claws back funding three months later.
Produces financial summaries and forecasts
How much funding are you expecting this term? How does it compare to last term? Where are the gaps between funded and actual costs? Your agent produces the financial picture your accountant wishes you already had.
The real numbers.
| Manager time on funding claims and tracking | £1,500–£3,000/year |
| Errors, late submissions, and clawbacks | £500–£2,000/year |
| Accountant or bookkeeper time reconciling funding | £300–£800/year |
| Realistic annual cost | £2,300–£5,800 |
| Agent build (one-off, configured to your LA's requirements) | £1,500–£2,500 |
| Monthly running costs (hosting + AI usage) | £50–£100/month |
| LA format and funding rate updates | Included in first year |
| Realistic first-year total | £2,100–£3,700 |
Funding claims shouldn't be this complicated, but they are. Multiple entitlement types, stretched and unstretched hours, eligibility codes that expire, LA-specific formats — and now termly head counts adding another cycle of returns.
The real cost isn't the manager's time (though that's significant). It's the money you leave on the table through under-claiming, or the money you have to return through over-claiming. Your agent doesn't make the funding system simpler — it makes your claims accurate.
Good fit / not a fit.
This works brilliantly for:
- Settings offering 30-hour extended entitlements with complex attendance patterns
- Nurseries that have experienced funding clawbacks due to claiming errors
- Multi-site operators managing claims across different local authorities
- Any setting where funding admin takes the manager more than a few hours per term
This probably isn't for you if:
- Your nursery management software handles funding claims end-to-end and you're satisfied with it
- You have a dedicated finance administrator who manages claims comfortably
- You only offer a small number of funded places and claims are straightforward
Handled like the sensitive data it is.
Children’s records — observations, safeguarding logs, SEND plans — are special category data under UK GDPR, and we treat them that way. Your agent runs on Claude via AWS Bedrock with an EU-only inference profile — meaning prompts and outputs never leave the EU, and are never used to train a model. Protected under Anthropic’s enterprise Business Associate Agreement.
Every run is logged for audit, every output is a draft your manager reviews and approves, and your data lives in a tenant isolated from every other customer. Parental consent verification is built into data workflows, and we’ll hand you a pre-filled DPIA template you can drop into your own records.
Need UK-only data residency? We offer an Azure UK South deployment as an enterprise add-on for customers with stricter procurement requirements. Full security details →
Let's talk.
We'll start with your funding setup — which entitlements you offer, how you track attendance, what your LA requires, and where the headaches usually are. Usually a 15-minute conversation.
hello@nimblecroft.com