When the health secretary Wes Streeting warned that parts of the NHS had “over-steered” towards financial control and were now under-delivering on elective activity, he named the problem precisely. The system has been managing by hindsight. Reports arrive late. Decisions follow later. Capacity leaks quietly in between.
Across Northwest London, a different model is running live. Less policy theatre. More systems engineering. Fewer static dashboards. More real-time control. And increasingly, AI used where it actually helps: prioritisation, validation and early warning.
This is not a pilot. It is production software, operating at scale.
From dashboards to operating systems
At Chelsea and Westminster Hospital NHS Foundation Trust, and across the wider Northwest London system, the NHS England Data and Digital programme has evolved beyond data aggregation into something closer to an operating system for care.
Nationally, the programme has been led by Ayub Bhayat, Director of Data Services and Deputy Chief Data and Analytics Officer at NHS England. His focus has been consistent: data platforms must deliver measurable clinical and operational impact, not just insight. The Federated Data Platform has been positioned as infrastructure, a shared data plane on which operational tools, automation and AI can run safely.
That direction is reinforced by executive sponsorship from Ming Tang, Interim Chief Digital and Information Officer and Chief Data and Analytics Officer at NHS England, and translated into delivery on the ground by Bruno Botelho, NWL Programme Director for the Federated Data Platform and Director of Digital Operations and Innovation at Chelsea and Westminster.
The alignment matters. Strategy defines the rules. Data provides the signal. Operations execute in real time.
Elective care, treated as a control problem
Elective recovery is where this architecture shows its sharpest edge.
Instead of managing theatres, waiting lists and cancellations through fragmented, manual, non-automated processes, teams now operate from a single, live view of demand and capacity. Theatre utilisation, list construction, RTT position and downstream flow are visible together, updated continuously rather than reconciled after the fact. This reframes elective care as a control problem rather than a reporting one.
AI-assisted logic is being used to validate waiting lists, flag risk, and support smarter prioritisation before pressure turns into cancellations or breaches. The effect is subtle but decisive. Fewer surprises. Tighter utilisation. Faster corrective action when the system starts to drift.
In practical terms, this has translated into materially higher theatre utilisation, more cases delivered per session and better use of existing capacity, without degrading cancellation performance. Latent capacity is being unlocked not by working harder, but by improving the signal quality leaders act on.
Cancer pathways as a stress test
Cancer services provide a tougher proving ground. Political scrutiny is intense. Clinical risk tolerance is low. Here too, the same digital foundations are changing outcomes. Multidisciplinary teams now operate from a shared, real-time view of referrals, diagnostics and pathway milestones. AI-supported validation and orchestration reduce administrative drag and surface delays earlier, when intervention still changes the outcome.
The result is counter-intuitive but repeatable. Higher referral volumes are being absorbed without expanding MDT coordination teams. Faster Diagnosis Standard performance is sustained. Clinicians spend less time chasing information and more time making decisions. This is not automation replacing judgement. It is augmentation removing noise.
Digital transformation that compounds
The financial signal follows the operational one.
At steady state, the programme is now releasing around £2m per year in recurring productivity and cost avoidance, delivered using existing staff and estate. Over a seven-year transformation horizon, consistent with major NHS operating model and digital programmes, this equates to approximately £14m of cumulative value. What matters is not the headline number. It is the mechanism. These gains compound because they are embedded into how the system runs day to day, not delivered as one-off recovery pushes.
Why this feels different
Most NHS digital programmes promise insight. This one deliver control.
Ayub Bhayat’s national leadership has framed data platforms as critical infrastructure for clinical and operational delivery. Ming Tang’s insistence on executive-level ownership has anchored the work at board level. Bruno Botelho’s delivery model has ensured the technology lands in real operational practice, not slideware.
In a system often characterised by long discussions and slow feedback loops, Northwest London is running something closer to a live, adaptive system.
As the NHS responds to Wes Streeting’s challenge to restore elective performance without losing financial discipline, this approach offers a glimpse of the future. Less retrospective reporting. More real-time control. AI used quietly, precisely, and where it works.
