Sequential Systems. Simultaneous Crises.
The workforce systems behind your public sector operations were built for a world that no longer exists.
Article 1 in the Connected or Exposed series
The world is a connected, multi-dimensional environment. Challenges are compound and spill over borders. These facts have always been true. What’s new is how many of these challenges are arriving at once, and how visibly the systems behind critical decisions can’t keep up.
Three patterns, right now. A military running sustained combat operations abroad while restructuring its civilian workforce back home: thousands of positions declared surplus, the medevac system stress-tested while the civilian workforce is being rebalanced at pace. Allied nations ramping up defence spending toward wartime levels, some with more than a fifth of military positions unfilled. Countries considering reinstating compulsory service for the first time in a generation. The recruitment pool is physically shrinking: births at a sixty-year low. A major economy where the nursing shortage has been directly linked to increased hospital mortality, workforce growth sustained only by international recruitment. Three sectors. Three continents. Same challenge: nobody could see what they actually had.
The instinct is to read these as separate problems: a military under operational pressure, an alliance scaling at pace, a healthcare system in demographic trouble. Fixable with the right policy, the right budget, the right hire. But underneath each one is the same operational pressure, a legitimate question: who is available, qualified, deployable, and whose absence won't break something else. Existing systems cannot answer fast enough.
It isn't a headcount problem, though headcount is tightening everywhere. It isn't a budget problem, though budgets are under pressure. It's a data architecture problem. The operating environment has shifted: simultaneous conflicts, alliance stress-tests, pandemic aftershocks, climate disaster surges, domestic restructuring at pace. Pressure that used to arrive in sequence now arrives concurrently. The workforce intelligence systems built to support operational decisions were designed for a world where you dealt with one crisis, recovered, then dealt with the next.
That world is gone.
The shift isn't temporary
The temptation in every one of those scenarios is to treat the pressure as a peak: a conflict that will end, a restructuring that will settle, a recruitment cycle that will catch up. Plan through it, staff around it, wait it out.
The evidence doesn't support that reading. The OECD Employment Outlook 2025 documents the structural picture across its member states: aging workforces, shrinking talent pipelines, skills mismatches accumulating faster than organisations are modelling them. That's the slow variable.
The fast variable is operational tempo. The EU Civil Protection Mechanism was activated 64 times in 2025, responding to the war in Ukraine, the conflict in the Middle East, wildfires across Europe, storms in Ireland, Cuba, Jamaica, Sri Lanka and Vietnam. That isn't a spike year. That's the new baseline.
New Zealand's emergency management agency acknowledged what most governments haven't said publicly: the system does not have the capacity or capability to deal with significant, widespread events that impact multiple regions at once. Not "might struggle." Does not have, in the present tense.
The organisations still planning for episodic disruption, one crisis at a time, recovery in between, are building workforce strategies for an operating environment that no longer exists. The question isn't whether you'll face compound pressure. It's whether your systems can see clearly enough to navigate it when you do.
The compound question
Every workforce system in public sector can answer simple questions. How many people we have, where they’re assigned, what's their grade. These are inventory questions, and most organisations can produce answers, eventually, from an HR platform or a reporting dashboard.
The question that breaks things is compound. Who is available, with the right skills and clearances, in the right jurisdiction, whose redeployment won't create a critical gap somewhere else, right now? That's not one query. It's a chain of conditions, each dependent on the last, spanning systems that were never designed to talk to each other.
This is the pattern in every scenario from the opening. The military restructuring thousands of civilian roles has an AI tool matching surplus personnel to vacancies. The question is whether any matching tool can account for what the data underneath it doesn't capture: which of those people are mission-critical, and what breaks if they move. Several nations are short on visibility into what capabilities they actually hold, where the gaps compound, and which ones are survivable.
It isn’t only a military problem. When Germany hosted Euro 2024, the federal police mounted their largest deployment since the Bundespolizei was founded in 1951: 22,000 officers daily across ten cities, 580 foreign officers from partner nations, coordinated through a dedicated centre in Neuss because no existing system could manage deployment across sixteen state police forces, federal agencies, and international partners simultaneously. Again, the question wasn’t only about the number of officers. It was whether any system could see across sixteen state forces and a dozen partner nations, who could deploy without leaving critical gaps somewhere else.
The problem isn't missing data. Almost every organisation discussed here has the data. It exists in HR systems, rostering tools, procurement platforms, clearance databases, training records, contractor management systems. The problem is that it exists in pieces, unconnected, in systems that answer their own questions competently and nobody else's.
What connected looks like
The fix isn't another dashboard. It isn't a bigger HR system or a better spreadsheet. It's a data layer that connects what already exists: people, skills, roles, certifications, assignments, contractor relationships, organisational dependencies, and the relationships between all of them, traversable in real time.
That's a specific architectural choice. It means modelling workforce data as a network of relationships. When someone asks "who can I deploy to this function without creating a gap elsewhere," the answer requires traversing multiple conditions simultaneously: skill match, security vetting, current assignment, geographic availability, contractual constraints. A connected data model does that natively. A collection of siloed systems requires someone to manually stitch the answer together, usually too slowly to be effective.
The organisations navigating compound pressure aren't the ones with the largest headcount or the most technology. They're the ones with enough visibility into their operational fabric to make fast decisions under uncertainty. They can see what they have, what depends on what, and what is impacted if something moves. That visibility is the dividing line.
This series calls it what it is. You're either connected, with clear sight across your people, capabilities, dependencies, and risk, or you're exposed. And you won't know which one you are until something breaks.
The AI question
The parallel conversation happening across every public sector system right now is AI. The budgets are moving. The pilots are launching. The expectation is that AI will close the gap: fewer people, more capability, faster decisions.
Some of that is real. But the domains where workforce intelligence matters most: combat medicine, disaster response, disease surveillance, aged care, nuclear safety, are domains where the work is irreducibly human. You cannot yet automate a medevac nurse. You cannot yet replace a disease surveillance specialist with a language model. The frontline capacity these systems depend on people. Specific people, with specific skills, in specific places. AI doesn't change that. What AI can do is help organisations see where those people are, what they're doing, and what cascading impact they have if they move. But only if it has something trustworthy to reason over.
AI applied to fragmented, siloed workforce data doesn't produce insight. It produces fluent, confident wrong answers at machine speed. An AI tool matching surplus personnel to vacancies can't account for mission-critical dependencies it can't see. An AI powered operations assistant answering "who's available for deployment" can't traverse clearance status, current assignment, and contractual constraints if those things live in three systems that don't talk to each other. The model isn't the problem. The data underneath it is.
The data foundation matters more than the model, and getting it wrong doesn't just waste a technology investment. In these domains, it puts people at risk.
What comes next
When the data layer changes, the questions change.
Instead of “how many people do we have”, it becomes “what happens if we move these three.”
Instead of discovering a gap after something breaks, you see it forming.
Instead of a three-day scramble across six systems to answer a deployment question, you get an answer that accounts for skills, clearances and dependencies in seconds.
The problems don’t get simpler, but the ability to see them clearly and adapt in real time does.
This is the first in a series of articles. Each one examines a specific dimension of the workforce intelligence gap: the inability to surge what you can't see, the contractor dependencies hiding in plain sight, the succession risks accumulating silently, the clearance data that's wrong more often than it's right, and the AI investment that's moving faster than the data foundation beneath it.
While the evidence is current and the patterns are global, the question underneath all of them is the same: can your organisation answer the compound operational questions that the current environment demands, in time for the answer to matter?
Most leaders can’t say what would fail if their top three specialists became unavailable simultaneously, not in a structured transition- but tomorrow. That's not a planning gap. That's exposure.