The Invisible Gatekeeper
Behind every bureaucratic victory or defeat, there is a screen. For thousands of people fleeing the wreckage of their former lives—families from Ukraine, individuals seeking safety from corners of the world most of us only see on the evening news—that screen is the only thing standing between a cold pavement and a roof.
For years, the British government relied on a digital architect named Palantir to build the scaffolding of the Homes for Ukraine scheme. Palantir is a name that carries weight. It evokes images of high-level intelligence, shadowy data mining, and the kind of "predictive policing" that makes civil liberties groups break out in a cold sweat. They are the titans of big data. But big data comes with a big price tag.
Recently, the Department for Levelling Up, Housing and Communities made a quiet, seismic shift. They walked away. They replaced the sophisticated, expensive software provided by the Silicon Valley giant with an in-house system.
The result? A saving of £2.3 million in a single year.
But this isn't just a story about spreadsheets and budget cuts. It is a story about what happens when a government decides it can build its own tools rather than renting its soul from a third party.
The Weight of a Digital Paper Trail
Imagine a woman named Olena. She is hypothetical, but her circumstances are lived by thousands. Olena arrives at a border with a folder of documents and a smartphone with a dying battery. To the system, she is a series of data points: a passport number, a background check, a match with a host family in a London suburb.
Under the old regime, the data flowed through the Foundry platform. It was efficient. It was powerful. It was also incredibly expensive to maintain. Every time the government wanted to tweak a parameter or pull a specific report, the meter was running.
When you outsource the "brain" of a humanitarian crisis to a private corporation, you aren't just buying software. You are buying a dependency. You are betting that the company’s algorithms will understand the nuance of a local council’s housing shortage or the specific needs of a traumatized child better than the civil servants on the ground.
The Department eventually realized that the "black box" approach wasn't just draining the coffers; it was distancing the makers of the policy from the machinery of its execution. By moving the refugee data system to an internal platform, the government didn't just save millions. They took back the keys to their own house.
The Myth of the Necessary Giant
We have been conditioned to believe that the public sector is inherently clunky, slow, and incapable of innovation. We assume that if a task is complex, we must hire a billion-dollar firm with a name that sounds like something out of a Tolkien novel to solve it.
This is a dangerous fallacy.
The transition away from Palantir’s tech proves that the "clunky" civil service can be nimble when the stakes are high enough. The new system handles the same heavy lifting: the "back-end" checks, the matching of sponsors, the tracking of payments. It does so without the astronomical licensing fees that typically characterize defense-grade software.
Consider the math. A saving of over £2 million isn't just a rounding error. In the context of local government, that is the salary of dozens of social workers. It is the funding for community centers. It is the difference between a refugee integration program that thrives and one that merely survives on fumes.
The logic used to be: "No one ever got fired for buying IBM." Today, that sentiment has shifted to the giants of data analytics. But the tide is turning. Governments are beginning to ask why they are paying a premium for features they don't use and data architectures they don't fully control.
When the Code Meets the Curb
Data is cold. A database doesn't feel the tension in a room when a host family and a refugee meet for the first time. It doesn't understand the anxiety of a visa delay.
However, the software shapes those experiences. If a system is designed for "intelligence and surveillance," it treats every user like a potential threat to be analyzed. If a system is designed by the people who actually run the program, it can be built with a different DNA. It can be built for service.
The shift away from Palantir in the refugee sector is a microcosm of a much larger struggle for digital sovereignty. For a long time, the UK government has been criticized for its "lock-in" with large tech providers. It’s like renting a car for a decade instead of learning how to drive. You eventually realize you've paid for the car five times over, and you still don't own the steering wheel.
By developing their own system for the refugee crisis, the DLUHC proved that the expertise exists within the walls of Whitehall. They stripped away the "bells and whistles" of a military-grade platform and built a tool that fits the hand of the person using it.
The Cost of the Human Element
We often talk about "millions saved" as if the money simply vanishes back into a vault. In reality, that money represents capacity.
When the government relies on an external provider, they are also exporting their problems. If the system crashes, they call a help desk in California or London. If the data is siloed, they wait for a consultant to bridge the gap.
Building internally is harder at first. It requires a level of accountability that you can't just outsource. But it creates a feedback loop. The people writing the code are sitting three desks away from the people talking to the local councils. They see the friction. They hear the complaints. They can fix the glitch in the afternoon rather than waiting for a contract renewal in the next quarter.
This isn't to say that Palantir doesn't have its uses. In the realms of national security or complex health data across millions of patients, their tools are undeniably powerful. But for a specific, human-centric mission like the Homes for Ukraine scheme, the "sledgehammer to crack a nut" approach was costing the taxpayer a fortune.
The Quiet Revolution
The departure from Palantir wasn't a loud, public breakup. It was a gradual migration. It was a realization that the "high-tech" solution isn't always the "right-tech" solution.
We are entering an era where the glamour of "Big Tech" is wearing off. We are seeing the bill, and it is staggering. As more departments look at their balance sheets, they are finding that the "innovation" they were promised is often just a very expensive way of doing what they could already do themselves.
There is a certain irony in the fact that a system designed to help people find homes was itself in need of a permanent place to stay—within the government's own infrastructure.
The invisible stakes here aren't just about pounds and pence. They are about the precedent. If we can build a system to manage one of the most significant refugee crises in modern European history without relying on the tech giants, what else can we do? Can we build our own health systems? Our own educational platforms? Our own social safety nets?
The Resonant Chord
As the sun sets over a small flat in Leeds, Olena sits at a kitchen table. She doesn't know about Palantir. She doesn't know about the £2.3 million saved by a group of developers in a government office. She only knows that her application was processed, her host was vetted, and she has a key in her hand.
The technology that brought her here has become invisible. And that is exactly how it should be. The best tools are the ones that disappear, leaving behind only the human connection they were built to facilitate.
The millions saved are a victory for the taxpayer, certainly. But the real triumph is the quiet realization that we don't need to be saved by a ghost in the machine. Sometimes, the most powerful thing a government can do is trust its own people to build the bridge.
The screen is still there. But the hand moving the mouse now belongs to someone who knows the name of the person on the other side of the data. That is a value that no license fee can ever truly cover.
