Treasury extended the SEND statutory override to March 2028. The councils that treat that as runway — not relief — will be the ones still standing on the other side.
For the last three years, every Section 151 Officer in England has been counting down to a date — March 2026 — when SEND deficits could no longer be parked off the balance sheet. The National Audit Office had warned that four in ten councils were at risk of issuing Section 114 notices if the override fell.
It hasn't fallen. The override now runs to March 2028.
That is, by any reading, a relief. But it is also the most dangerous kind of relief: the kind that gives leadership teams permission to assume the problem will be solved by someone else, in some other budget cycle, by some other minister.
The numbers haven't moved. The cumulative SEND deficit across English councils sits in the region of £5 billion and continues to grow. CIPFA modelling has it pushing past £8 billion in 2026/27 if current trajectories hold.
Moreover, while it doesn’t appear on the balance sheet, the costs of managing that deficit are still significant – be it lost income from not investing or the actual costs of issuing debt – which can run into the millions. A promise to pay off 90% of the current deficit is welcome, but doesn’t solve the ongoing deficit – and payment is contingent on creating a more financially sustainable service.
Two extra years is not a fix. It is a window. And what councils do inside that window is what separates the authorities that will look financially sustainable in 2028 from the ones that will be back at the same crossroads, with the same headlines and the same options narrowed by another political cycle.
So here is the operational question that I think every Chief Executive, CIO, Section 151 and Director of Children's Services should be asking, openly, this quarter: how much of our 2027/28 plan is being funded by digitally-released value, and how much is being funded by hope that the override gets extended again?
Hope is not a strategy any of us would write down as a strategy. But it is the strategy a lot of councils are quietly running, because the alternative — using the next 24 months to release real value out of SEND, social care and the corporate operating model — has felt too slow to start.
It isn't slow anymore.
Three shifts in the last 12 months mean a council can now release SEND-relevant value in weeks rather than years.
The first is AI in casework. Kingston Council's adult social care pilot has shown that an AI-supported workflow can take meaningful time off the highest-volume parts of an assessment — case notes, triage summaries, drafting — without removing professional judgement. The same shape of intervention applies, almost cleanly, to EHCP drafting and annual reviews, where the bottleneck is precisely the same: a small number of statutory officers writing high volumes of structured prose. This is not a moonshot. It is a redeployment of hours that already exist.
The second is the front door. SEND demand is shaped, not fixed. Councils that triage at the point of first contact — clear eligibility, clear pathways, clear early-help offers — see far fewer cases escalate to statutory assessment than councils that route everything through the same single funnel. Modern integration platforms, low-code workflows and decision-support models make this redesign possible without ripping out a casework system. The change is in the routing, not the tools.
The third is supply. The market for SEND placements, particularly independent specialist provision, has been dysfunctional for a decade — and the cost release sits there. Councils that have moved to a pre-assessed, marketplace approach to specialist provision — with quality, distance, and unit cost visible at the point of placement decision — have demonstrably bent their high-needs spend curve in months, not years.
Technologies like this have a solid and growing evidence base in addressing SEN deficits.
None of this is a critique of the partners councils have traditionally leaned on. Many of the big firms have done excellent SEND review work, and continue to. The shift isn't quality of advice. It is timing. Reviews that arrive in nine months are arriving inside the same 24-month window the council itself is racing against. Different shape, different product, different moment.
The triple lens is unforgiving on SEND, in a useful way. Outcomes: are children getting the right support at the right time? Experiences: do families feel the difference, or are they still grinding through processes designed around the council's convenience? Efficiencies: is every pound of high-needs block returning a quantifiable, sustainable, positive outcome — or is it being absorbed into a system that has lost its ability to evidence value, even facilitating dependence in SEN children?
If the honest answer to any of those three is "no", then the override extension is not your safety net. It is your stopwatch.
I'd gently challenge anyone reading this with a SEND portfolio: take your high-needs financial recovery plan, mark every line that depends on a future grant, a future reform, or a future override extension, and look at what is left. That is the part of the plan that is actually within your control. That is the part that digital can move.
The councils I am watching in 2026 — Walsall, Denbighshire, the three Cambridgeshire authorities, and others I won't name yet — are doing exactly that. They aren't waiting for a national settlement. They are using a 1,300-process baseline of where local government work happens, choosing a tightly-scoped 40-opportunity shortlist, and moving the highest-leverage SEND and social care plays into delivery while the override clock is still on their side.
That is what runway looks like. The relief is real, but the work is now.
I'd like to hear from anyone leading SEND or children's services right now: what is the single intervention you'd unlock in the next 18 months if the right partner walked into the room with the case already half-built? The answers in the comments tend to be where the next 24 months start.
