Some problems resist analysis the way oil resists water. You apply more rigour, sharper frameworks, cleaner data. The problem remains. Another meeting, another deck, another working group. The problem remains. At some point the suspicion arrives — quiet, uncomfortable — that thinking harder is not the issue. That the problem is not what it appears to be.
Most organisations have a name for this. They call it a complex problem, or an alignment problem, or a resourcing problem. The name keeps changing because the analysis keeps producing the right answer and the right answer keeps not being implemented.
What is actually happening, in most of these cases, is that the problem is political. And political problems do not respond to analytical solutions.
The distinction matters more than most people want it to.
An analytical problem has a best answer that, once found, most reasonable people will recognise and act on. The work is in the finding. Once found, the finding does the rest.
A political problem has a best answer that is already known — or at least knowable — and is not being acted on because acting on it would shift power, redistribute resources, threaten a relationship, require someone influential to admit they were wrong, or make visible a dynamic that has been profitable to keep invisible. The work is not in the finding. The analysis is done. What remains is the human system resisting the implication.
Applying more analysis to a political problem does not solve it. It produces more evidence of the answer that cannot yet be acted on — and sometimes makes the resistance harder, because the case is now clearer and the people resisting it must work harder to justify the position.
The diagnosis is harder than it sounds.
Political problems are almost never presented as political. They arrive wearing analytical clothes — framed as questions of data quality, strategic uncertainty, insufficient stakeholder input, or the need for further discovery. Each of these framings is occasionally true. They are also the most common ways that organisations avoid the conversation that would actually resolve the problem.
The signal that you are in a political problem is not the absence of a good answer. It is the pattern of what happens to good answers once they arrive.
If the analysis is received, acknowledged, and then quietly set aside — the problem is political. If the same question has been studied three times by three different teams and each time produced a similar conclusion that went nowhere — the problem is political. If the people who know the answer are not the people who need to act on it, and the people who need to act on it have reasons not to — the problem is political.
The question to ask is not what does the data say. It is who would be affected if the data were acted on, and whether those people have more influence than the data.
The mistake is a category error, and it is understandable.
Most people in analytical roles were trained, promoted, and rewarded for solving analytical problems. The instrument works. When it stops working, the instinct is to sharpen it — better analysis, more robust methodology, wider consultation. This is the right response to an analytical problem. It is the wrong response to a political one, and it costs time, credibility, and the patience of the people who already knew the answer.
Recognising the category error early — before the third study is commissioned, before the fourth working group is formed — is one of the more valuable things a senior operator can do. Not because it saves the analysis. Because it forces the real question into the room: if we already know what needs to happen, what is actually preventing it?
That question is uncomfortable. It names that the obstacle is not ignorance but interest — not a gap in understanding but a gap between what is right and what is convenient for the people who have the power to act. This is the conversation that organisations are structured, in many ways, to avoid. The leader who names it anyway — precisely, without drama, with the evidence visible — is doing something that no amount of additional analysis could do.
Political problems require different tools.
Not the absence of analysis — the analysis is often still needed, as the factual foundation for what comes next. But the work, once the analysis is done, shifts from intellectual to relational and structural.
The first move is to identify who has the most to lose from resolution. Not who disagrees with the analysis — that is a list of the wrong people to spend time on. Who has a material interest in the problem remaining unsolved? Their position is not irrational. It is self-interested in a way that analysis cannot overcome. Understanding it clearly is the prerequisite for knowing what would have to change for them to move.
The second is to find the person with both the authority and the incentive to act. Political problems usually resolve when someone with enough power decides the cost of the status quo is higher than the cost of the disruption. Your job is often to make the cost of the status quo visible to that person — specifically, in terms of what it is costing them, not in terms of what it is costing the org in the abstract.
The third is to change the conditions rather than the argument. If the argument has not worked three times, a fourth iteration of the argument will not work either. What changes is the context — new information that reframes the stakes, a shift in the external environment that makes the status quo suddenly expensive, a structural change that alters who benefits from resolution. You are not persuading. You are waiting for or creating the conditions under which the political calculus shifts.
The fourth is to name it — carefully, to the right person, once. Not “this is a political problem” — that is a conversation-ender in most orgs. Something more precise: “The analysis has been done. The answer is known. I want to understand what would need to be true for us to act on it.” This is not naive. It is a way of moving the conversation from the question of what is right to the question of what is actually blocking movement — which is where the real work is.
The hardest part is not the diagnosis. It is accepting that the solution is not in your hands.
The operator who identifies a political problem clearly, surfaces it precisely, and creates the conditions for movement has done everything within their scope. Whether the system acts on it is a function of forces — interests, relationships, histories, incentives — that are largely outside anyone’s individual control. This is frustrating. It is also accurate.
The failure mode is continuing to treat a political problem as analytical because the alternative — that the solution depends on the system being willing to change, not on the analysis being better — is harder to accept. The analytical person has agency. The political reality requires something closer to patience, strategic positioning, and the willingness to plant the right question and wait for the conditions that make it answerable.
When the answer is known and nothing moves, stop asking what the data says. Start asking who needs the problem to stay unsolved — and what would have to change for them to stop needing that.