MPs heard Manchester is better than FDP. It seems they were misled

What Greater Manchester has

Greater Manchester's Advanced Data Science Platform is a data warehouse. It is built on Microsoft SQL Server, with Tableau as the visualisation layer, Graphnet's CareCentric as the shared care record, and Snowflake as the analytics environment. It links datasets across acute, primary care, community, mental health and social care settings, covering a population of 2.8 million. It supports population health management, risk stratification, service planning and research through its Secure Data Environment. These are legitimate capabilities that took a decade and, at a conservative estimate, £60m to £80m of combined programme funding, staff costs and infrastructure investment to build.

They are also, in architectural terms, unremarkable. Dozens of ICBs and large trusts across England have data warehouses with linked datasets, visualisation tools and shared care records. Hennessey named several at the committee session: Leeds, Dorset, Kent. Wrigley added OneLondon and Thames Valley. What neither of them acknowledged is that each of these platforms was built independently, on a different technology stack, against a different data model, by a different team. Leeds runs on a different architecture to Dorset. OneLondon is built differently again. They do not connect to each other, and they were never designed to.  

Wrigley's answer to this fragmentation is interoperability: connect the existing platforms through open standards and the problem is solved. This sounds reasonable in principle, and may well work in the comparatively simple data domain of mobile phone calls - Wrigley's experience - where data is simple and standards were created at the outset. In practice at the NHS, it would mean retrofitting and maintaining bilateral agreements, semantic mappings and integration pipelines across dozens of organisations, each running different database technologies, different visualisation tools, different shared care record suppliers and different data models. Even a small change to such a platform at one organisation would need endless governance to ensure the system as a whole still worked, with an ever-present risk of catastrophic error. The NHS has been attempting interoperability between its existing systems for over twenty years. The reason FDP exists is that it has not worked. 

Greater Manchester's advantage is not architectural novelty. It is time: devolution in 2016 gave GM a head start and sustained political support for investment in data infrastructure that other ICBs did not have. At that time, GM ICB’s data capability was described as "a nest of spreadsheets." What the team built from that starting point is a well-executed version of a common pattern. A data warehouse with analytics and a shared care record. Not a platform that manages clinical workflows at the point of care.

This distinction is critical, and it is the one the select committee failed to examine.

What FDP has that Manchester does not

FDP's core products include the inpatient care coordination solution, RTT Validation, Cancer360, and the discharge planning products now in deployment. In the Trusts where these are live, clinical staff use them on wards to manage patient pathways, identify patients who have been waiting longest, schedule and optimise theatre sessions, coordinate discharges across teams, and track cancer patients through diagnostic and treatment milestones. These products do not sit in a reporting suite where an analyst produces a dashboard for a board meeting. They sit in the hands of ward coordinators, theatre managers, discharge teams and cancer pathway navigators. Staff take actions through these FDP products: flagging a patient for review, booking a theatre slot, escalating a delayed discharge. The products generate new structured data as a byproduct of those actions, data that feeds back into the same data fabric and becomes available for analytics, reporting and research.

This is the concept I have called a Frontline First approach throughout the “Inside FDP” series. The platform replaces the spreadsheets, emails, whiteboards and paper trackers that clinical teams currently use to coordinate care. Those shadow systems are invisible to information governance frameworks and produce no accessible structured data. Not only do these represent an information governance risk, these informal systems are cited in patient safety inquiry after patient safety inquiry, not least the recent Ockenden and Amos reports in maternity.  

Some forward-thinking trusts are already building their own tools on FDP to address these issues. As one group of hospital trusts told the Kings Fund recently, FDP can be used by clinical teams themselves to replace their workarounds with governed, auditable, nationally consistent digital products. This allows clinicians for the first time to actively address the risks they already know about but have been prevented from acting on due to lack of access to the right systems. The vision of these trusts is to move away from IT gatekeeping to create an ecosystem of "citizen developers" or "colleague developers" who can build their own custom tools on top of high-quality, governed data. 

FDP also operates a national instance that no regional platform can replicate. This is the layer that allows NHS England and the DHSC to stand up nationwide public health programmes, including COVID, flu, MMR and meningitis B vaccination campaigns, in a matter of weeks rather than months. National orchestration requires national infrastructure. Greater Manchester's platform, by design, serves one region.

Greater Manchester ICB’s data warehouse does not do any of this. However well linked, however well visualised, however wide the dataset, a warehouse sits alongside the operational workflow. It tells you what happened. It does not put a tool in the hands of a theatre manager to rearrange tomorrow's list, or in the hands of a discharge coordinator to see which patients across three wards are ready to leave and which are blocked by a delayed social care referral. A data warehouse tells you what happened. An operational platform tells a clinician what to do next. These are not the same thing, and no amount of parliamentary testimony makes them the same thing. 

The ICB question 

There is a further dimension that Hennessey's framing obscures. FDP is not limited to Trust-level products. It includes products designed specifically for ICBs: the Strategic Commissioning Tool, which provides population segmentation, resource allocation modelling and benchmarking against other ICBs nationally; and the System Co-ordination Centre, which integrates real-time data across all providers within an ICB for demand and capacity management. These are commissioning products, built for the strategic commissioner tier that Hennessey described as GM's domain. The difference is that they operate against a national data model and allow national benchmarking, which a standalone regional platform cannot deliver because there is nothing to benchmark against. Even on Hennessey's own terms, FDP serves both tiers. GM's ADSP serves one. 

On his own patch, Hennessey has a point. GM's platform, refined over a decade against local requirements, and connected up with primary care, is almost certainly better suited to GM's commissioning needs than the FDP's ICB products would be today. A generic national product will rarely outperform a purpose-built local system with ten years of development behind it. I have spoken to operational staff at other ICBs with mature local systems who report the same pattern: their existing infrastructure is, for now, ahead of what FDP's ICB products can offer in their geography. But Hennessey's position rests on generalising from one ICB's exceptional circumstances to a national policy conclusion. The reason FDP's ICB products exist is that 35 other ICBs did not receive the investment, the devolution settlement, or the decade of runway that GM had. For those organisations, a nationally provided commissioning and coordination toolset is not a downgrade from what they have. For many of them, it is the first time they have had anything beyond a spreadsheet and a Tableau licence.

The contradiction

Now to the evidence session. Matt Hennessey, representing the Chief Data and Analytical Officer Network, gave the committee a technically accurate account of the distinction. He told them: "It is important to make a distinction between what FDP offers ICBs and what it offers trusts. The infrastructure in the stack that has been built for Greater Manchester was built specifically to meet the needs of a strategic commissioner and the role that the ICB has." When asked about GM's inclusion of GP data, he confirmed it, but added that this was "distinctly different from an operational workflow that might be in a trust." 

He was right. And then, twelve minutes later, he said the opposite. 

Wrigley asked directly how much of FDP's capability Greater Manchester already has. Hennessey responded that "the technology that we have can match anything that the Palantir FDP can do." He said GM's functionality was "greater" because it includes GP data, adult social care data and local flows. Wrigley then summarised: "you have everything you need, you have everything the FDP can do, and more, because you have the single patient record already existing across Greater Manchester?" Hennessey agreed.

What followed was an escalation. Each of Wrigley's subsequent questions built on the previous agreement, extending the claim further than the witness was willing to go. When Wrigley asserted that around 30% of the population was already covered by similar systems, Hennessey pulled back: "I would not say to what percentage."

When Wrigley claimed these systems were "already providing benefits arguably greater than Palantir" and covered "more trusts and more people than Palantir does today," Hennessey declined again: "I cannot validate the numbers you talk about." But each refusal was followed by a partial concession ("but yes, there is definitely an existing infrastructure"), and Wrigley treated each partial concession as full confirmation before building the next question on top of it. By question 41, the claim on the record had expanded to cover "sophisticated technology doing all of the things that we want to do in terms of federated data, data analysis and the single patient record already provided by UK-based companies." Hennessey never endorsed that summary. Wrigley never needed him to. The cumulative weight of five partial agreements created a record that says something the witness never actually said.

These two positions cannot both be true. If GM's infrastructure was "built specifically to meet the needs of a strategic commissioner" and is "distinctly different from an operational workflow," then it cannot "match anything" FDP does. The first describes a data warehouse built for commissioning analytics. The second describes a platform that includes the operational clinical products described above. Hennessey drew the line himself, and then immediately erased it.

The most damaging evidence against his claim came from his own region. In the second panel, Ayub Bhayat confirmed that FDP's single-queue diagnostic product is "live across every single hospital in Greater Manchester. Every NHS trust in Greater Manchester uses a single queue diagnostic tool." The ICB refused the platform while the acute Trusts adopted it. Within Greater Manchester itself, the hospitals that manage patients on wards chose FDP's operational products over the ADSP for the clinical workflows that Wrigley was asking about. The witness was telling the committee that GM's platform can match anything FDP does. The hospitals in his own patch had already concluded otherwise.

The committee did not want to hear what FDP does

The select committee session was not a neutral examination of evidence. Martin Wrigley MP was there as a guest of the committee. The oral evidence was presented by three people who have never logged in to FDP and are nowhere close to the detail. The Minister, Preet Kaur Gill, attempted on multiple occasions to explain what FDP's operational products do at the clinical frontline. She described clinicians making "day-to-day decisions about cancer pathways, how many beds are available, theatre utilisation, discharges." She described wards replacing whiteboards with digital systems. She was describing the Frontline First layer, the operational products that distinguish FDP from a data warehouse.

The committee stopped her multiple times.

Paulette Hamilton cut her off at Q60: "I am not asking you what it is, Minister Gill." Ben Coleman interrupted at Q67 with the Chair's support: "minister, can I please urge you to directly answer the questions as they are presented." Coleman cut her off again at Q83: "No, that was not a question about horizon scanning." The Chair, Layla Moran, intervened at Q110: "That was not my question," forcing the minister to ask "What was your question?" And Wrigley himself stopped her at Q92 as she tried to describe the clinical benefits: "Forgive me, minister, but your time is tight, so we do not need to go through the benefits to the patient. We understand that."

A select committee scrutinising a clinical data platform told the minister they did not need to hear about the benefits to patients. Let that sink in.

The result was a session where the committee heard extensive testimony about analytics capability from a witness who conflated it with operational products, and then actively prevented the Minister from explaining the difference. The operational distinction that is central to evaluating whether alternatives to FDP exist was not examined because the one person trying to explain it was repeatedly told to stop.

A campaign, not an inquiry

The select committee exchange did not emerge from nowhere. The timeline is documented and public.

In February 2025, the Chief Data and Analytical Officer Network (CDAON) published an open letter claiming that ICBs "already have similar tools in use that presently exceed the capability and application of what the FDP is currently trying to develop or roll out at a system level." Hennessey was instrumental in that letter. In May 2025, his board paper to GM ICB described FDP adoption as "a time-consuming and possibly retrograde step." GM became the only ICB in England to formally refuse the platform. 

In March 2026, the campaign organisation Foxglove co-hosted what it described as a "drop-in session" in Westminster with Wrigley, inviting MPs and peers to hear briefings about Palantir. Foxglove described the purpose as building "a caucus of lawmakers in the UK parliament large enough to help convince the government that Palantir's contract for the FDP must end." Supporting campaign organisations included Medact, the Good Law Project, Privacy International, Just Treatment, Corporate Watch and the United Tech and Allied Workers Union.

In April 2026, Wrigley secured and led a Westminster Hall debate on FDP. Only a dozen or so MPs attended, and every one who spoke raised concerns about Palantir. The CDAON letter was cited as evidence. Hennessey posted on LinkedIn that he followed the debate "with considerable interest.". This “debate” was amplified in the left wing media with headlines like "Labour and Lib Dem MPs demand 'shameful' Palantir NHS contract be scrapped".

In May 2026, Medact published a briefing that synthesised the CDAON letter, the GM ICB board papers and the campaign's wider arguments into a single document, endorsed by the same coalition. GM ICB confirmed through an FOI response that it would not review its refusal to adopt FDP, citing "heightened public concerns."

On 3 June 2026, the Science, Innovation and Technology Committee recommended using the break clause to end the Palantir contract.

On 16 June, Hennessey appeared before the Health and Social Care Committee. Wrigley sat on the committee as a guest. His questions at Q35 through Q41 were constructed to produce the specific claims his off-ramp argument requires. On 7 July, the Canary reported Wrigley is now building that off-ramp, with Long-Bailey citing Greater Manchester as the model.

The questions Wrigley asked were not designed to test whether GM's platform can replace FDP's operational products. They were designed to get confirmation that it can, from a witness whose own published work provided the campaign's primary evidence base. The campaign chose the question, chose the questioner and chose the witness. The committee obliged by preventing the Minister from complicating the answer.

The committee's letter to the Minister, published on 8 July, extends this pattern. It cites the Minister's Q95 answer, in which she confirmed that Trusts with good existing systems would not be forced to adopt FDP, as evidence that 'the Government accepts that better tools exist.' It then uses this reframing to recommend contract termination. But the Minister's statement was a concession about local flexibility, not an acceptance that existing platforms can replace FDP nationally. The committee has converted one into the other. 

What it would cost and what it would break

If Greater Manchester's platform is the model for a national sovereign alternative, two practical questions follow.

The first is cost. GM's traceable programme funding, across Connected Health Cities, the Local Health and Care Record Exemplar programme, GMCA co-investment, the digital platform supplier suite and the Secure Data Environment, totals approximately £36 million. With a decade of staff costs, ongoing licensing for Graphnet, Tableau and Snowflake, and cloud infrastructure, the total is conservatively £60 to £80 million. For one ICB. Replicating this infrastructure at ICB level 36 times, even before building the operational products that GM does not have, across 220 NHS Trusts, is not a policy that any MP advocating the off-ramp has costed.

And cost is only half the problem. GM had a decade of runway, devolution-era political support, and a mature local team to draw on. Many ICBs were formed in 2022, lack specialist data engineering capacity, and are currently absorbing 50% headcount cuts to their running cost allowances. Proposing that these organisations replicate a bespoke ten-year build while simultaneously losing staff sounds more like a slogan than a plan.

The second is transition. Rob Thompson, the most senior digital official in both the Department of Health and NHS England, told the second panel that if the Government chose to exit the contract, "the benefits will stop as we exit." Development on the existing platform would have to halt "to give the new platform time to catch up." The NHS would be "moving from a supplier that has a high-performance platform, it works, and it works incredibly well, and we will be taking on trust that the alternative will work as well." The original FDP procurement took over 15 months. Bhayat confirmed that December is the latest possible date for triggering the break clause. He could not confirm that a new provider would be in place by March 2027.

The 139 trusts currently live on FDP, reporting measurable benefits from its operational products, would face a period of unknown duration during which the products they use daily to manage patient flow, schedule theatres and coordinate discharges would be withdrawn, with no guaranteed replacement ready. Those trusts would go back to the spreadsheets, emails and whiteboards that FDP was designed to replace. Back to the shadow systems that no trust has audited, that no trust has on its Board Assurance Framework, and that carry clinical safety risk, sometimes fatal, every day they remain in use.

The question that should have been asked

The decision on whether to exercise the break clause should be based on evidence that has been tested, not on a claim that contradicted the witness's own earlier testimony and was never challenged. If MPs are serious about scrutinising this contract, they should ask the question that was not asked on 16 June. 

Can a ward coordinator at Manchester University NHS Foundation Trust open the ADSP and manage a theatre list? Can a discharge team at Salford Royal use it to coordinate discharges across wards? Can a cancer pathway navigator at The Christie track a patient through diagnostics, staging and treatment milestones using products on the Greater Manchester platform? 

If the answer is yes, the off-ramp has a destination. If the answer is no, then Parliament is preparing to cancel a £330m contract, withdraw operational products from 139 Trusts and send clinical teams back to spreadsheets, on the basis of a false equivalence that a single line of enquiry would have exposed, had the committee members been truly curious.

Tom Bartlett is the founder of Bartlett Data Ltd and former deputy director of data engineering at NHS England, where he led the approximately 150-person engineering team that built the national FDP products.