The Pressure Map

Polis combines five public service indicators — unemployment, food bank usage, insolvency rates, crime, and NHS waiting times — into a single service pressure score for every constituency in England and Wales. It then measures how much online political engagement each constituency generates through parliamentary petitions.

The gap between these two numbers reveals something uncomfortable about British democracy: the places under the most material strain are often the least visible in the online systems that increasingly shape political attention.

One Seat

Birmingham Ladywood. The Lord Chancellor's constituency. Third-highest service pressure in England. And one of the quietest places online in the country.

Those four numbers tell a story worth examining. A constituency where the composite weight of unemployment, food bank usage, crime, and NHS waiting times ranks among the worst in England — yet the residents barely register in the online political systems that increasingly shape where attention goes.

Ladywood is not alone.

The Pattern

When you plot every constituency's service pressure against its online engagement, a cluster emerges in the bottom-right corner: high pressure, low engagement. We call it the silent-pressure belt.

If online engagement tracked material conditions — if the constituencies under the most pressure were also the loudest online — the scatter would follow a rough diagonal. It does not.

Thirty-eight constituencies have service pressure above 50 and engagement below 30. That is nearly 6% of all seats in the House of Commons, effectively invisible to the digital monitoring systems that increasingly shape political attention.

The places under the most pressure are producing the least online signal. That is not a coincidence — it is a structural feature of British political geography.

The Funnel

Look at how that number narrows.

The 38 constituencies share a demographic profile that goes some way to explaining the pattern. They are overwhelmingly urban, disproportionately diverse, and skew younger and lower-income than the national average. These are populations with lower broadband penetration, lower digital literacy rates, and — critically — lower rates of the kind of civic-digital behaviour that petition platforms capture.

This does not mean these communities are politically disengaged. It means they are disengaged from the specific form of politics that generates a trackable digital signal.

The party composition raises its own questions. Of the 38, twenty-nine are held by Labour, four by Labour (Co-op), two by Independents, and one each by the Green Party, Your Party, and the Conservatives. The single Conservative seat — Chingford and Woodford Green, held by Sir Iain Duncan Smith — is the only one where the governing party faces meaningful electoral competition.

Why does this matter? Safe seats generate less competitive political activity. Less competition means less campaign infrastructure, less media attention, less pressure on the MP to be visibly responsive. Labour holds more urban seats with these characteristics, so Labour seats dominate the list. The mechanism is the safety of the majority itself.

The Cluster

Birmingham offers the clearest single-city illustration. All nine of its constituencies have service pressure above the national average of 31.4. Six are in the silent-pressure belt. The city does not merely appear in the pattern — it anchors it.

At the other end of the range, Northfield scores 56 — still well above the national average — but with an engagement gap of 50, matching the national norm. It is the only Birmingham seat where online engagement roughly corresponds to the pressure residents face. It also happens to be the Birmingham seat that was most competitive at the last general election. Draw your own conclusion.

Every Birmingham seat scores 95 on the reality gap. That means across the entire city, regardless of which MP holds the seat, the pattern of parliamentary questions, divisions, and early day motions bears almost no resemblance to the pattern of pressures their constituents face.

The accountability scores tell a more varied story. They range from 26.7 in Ladywood to 66.3 in Northfield. Where there is some competitive pressure and some online voice, the MP appears to respond more closely to local conditions. Where both are absent — as in Ladywood — the feedback loop between constituent need and parliamentary response appears to have broken down. The pattern repeats across the other urban clusters in the belt.

The Contrast

The silent-pressure belt becomes more intelligible when placed alongside constituencies where people do shout — where online engagement runs high regardless of pressure.

Clacton's engagement score of 98 does not mean its residents face greater hardship than Ladywood's. By the service pressure measure, they face considerably less. It means they are far more likely to express their political preferences through the specific digital channels we measure.

The demographic profiles are instructive. The constituencies with the highest online engagement — Clacton, Blackpool North and Fleetwood, Brentwood and Ongar — tend to be older, whiter, more likely to own property, and more digitally active in the specific ways that petition platforms capture. Clacton's extraordinary engagement score likely reflects a combination of factors: a constituency with unusually high political mobilisation, a high-profile MP whose national platform amplifies local engagement, and a demographic base that skews towards the petition-signing behaviours our index captures.

The gap between Ladywood's engagement score of 3 and Clacton's 98 is not a gap in political feeling. It is a gap in the mode of political expression, shaped by demographics, digital access, mobilisation infrastructure, and the cultural norms of how different communities interact with online political tools.

Clacton's engagement score is 98. Birmingham Ladywood's is 3. The difference is not in how much residents care about politics — it is in whether their caring produces a signal that digital political operations can detect.

Birkenhead offers a useful middle case. Its service pressure of 78 ranks it fourth nationally. But its engagement gap of 63 places it above the national average — pressured and moderately vocal. It falls outside the silent-pressure belt because, unlike Ladywood or Bradford West, its residents generate enough digital signal to register. Whether that signal translates into different MP behaviour is a question the reality gap and accountability scores can begin to address — but the signal itself exists. In the attention economy of digital politics, existing is the prerequisite for being heard.

The Stress Test

Any methodology worth its name should be tested against its most inconvenient output. For the pressure map, that is the Prime Minister's own constituency.

The reality gap of 95 — identical to every Birmingham seat — indicates that the domain distribution of the PM's parliamentary activity bears almost no resemblance to the pressures facing his constituents. His accountability score of 6.1 is the lowest in the entire silent-pressure belt.

There is an obvious and partially valid defence. A Prime Minister cannot reasonably be expected to operate as a constituency MP in the conventional sense. The demands of office redirect parliamentary activity towards national priorities. But the score does not measure effort or intent — it measures alignment between constituent need and parliamentary output, regardless of the reason for misalignment.

If the methodology produces uncomfortable results for the most powerful person in the dataset, and the workings are transparent, that is evidence of rigour rather than bias. A system that exempted the Prime Minister's seat from measurement would be the one worth questioning.

What This Means

The pressure map reveals a structural asymmetry in British political attention. The constituencies facing the most material pressure are disproportionately the ones generating the least online political signal.

Digital-first political monitoring is systematically biased. Any organisation that relies on online engagement metrics alone as a proxy for political salience will systematically underweight the constituencies in this belt. This is precisely the gap that Polis is designed to expose: by cross-referencing online engagement against offline service pressure, parliamentary activity, and accountability metrics, we surface the places that a single-signal model would miss entirely.

The reality gap is nearly universal in high-pressure urban seats. Every Birmingham constituency scores 95. The PM's seat scores 95. This suggests the disconnect between constituent need and parliamentary activity is not a function of individual MP performance but of how the parliamentary system distributes attention. MPs in safe seats face less competitive pressure. MPs with ministerial responsibilities face different demands. The result is the same.

The accountability score shows meaningful variation within the pattern. Accountability ranges from 6.1 (Holborn and St Pancras) to 66.3 (Birmingham Northfield). Some MPs do a materially better job of aligning their activity with constituent conditions, even within the structural constraints of safe seats and ministerial office. The variation is the finding: it means the gap is not inevitable.

The cross-reference between these four metrics tells a more complete story than any single measure. A constituency with high pressure, low engagement, high reality gap, and low accountability is one where every feedback mechanism has failed simultaneously. A constituency where even one of those metrics breaks the pattern suggests the system is not entirely closed. The difference matters — and it is only visible when multiple signals are considered together.

What's Missing

No analysis built on composite indices should be presented without an honest accounting of what those indices cannot capture.

Petition data is a narrow slice of political engagement. People write to their MPs, attend surgeries, organise locally, and vote. None of this is captured by the engagement gap score. A constituency scoring 1 may have a highly engaged population that simply does not use parliamentary petitions as its mode of political expression. The score measures digital petition activity, not political engagement writ large. We use it because it is the measurable, constituency-level, publicly available proxy — not because it is the best measure that could theoretically exist.

Service pressure is a relative composite, not an absolute measure of hardship. A score of 81 means Birmingham Ladywood faces more pressure than 98% of constituencies on our weighted combination of indicators. It does not mean 81% of residents are struggling. Different weighting choices would produce different rankings. We publish the weights so that readers can assess whether they match their own analytical framework.

The reality gap measures domain distribution, not quality. A score of 95 means the policy domains in which an MP is active (defence, trade, constitutional affairs) are very different from those in which their constituency faces pressure (employment, crime, health). It does not measure whether the MP's activity is effective within the domains they do address. An MP who asks excellent questions about trade policy but none about the unemployment crisis in their constituency will score a high reality gap. That is a limitation of the metric, but it is also its purpose: to measure alignment of attention, not quality of action.

Correlation is not causation. We observe that safe seats tend to have higher pressure and lower engagement simultaneously. We do not claim that safety causes inattention or that inattention causes safety. The causal chain is likely bidirectional and mediated by factors a composite index cannot disentangle.

The data has currency limits. Unemployment figures lag by a quarter. Food bank data is annual. Parliamentary activity is historical. The map is a snapshot, not a forecast.

What the analysis does show — with reasonable confidence — is that the geography of material pressure and the geography of digital political voice in the UK are substantially misaligned, and that this misalignment follows predictable patterns related to seat safety, demographics, and urban geography. Whether this matters depends on what you use the data for. For anyone whose work involves understanding where political pressure is building in Britain, it matters a great deal.

How We Measured It

This analysis was produced by Polis Intelligence using public data sources. The methodology and data are available for scrutiny. Constituency-level data can be explored on the Polis dashboard.