For years the story about crime in big cities ran in one direction, and it was never a good one. So it is worth pausing on a set of numbers that point the other way. Across 2025, Greater Manchester recorded something in the region of 10,000 fewer crimes than the year before. Burglary fell by more than a fifth. Robbery and knife crime both dropped. The force solved roughly 47,500 offences, around 15 per cent more than the previous year, and it now answers 999 calls in an average of about five seconds. Independent Office for National Statistics figures told a similar story, with burglary, robbery and theft from the person all falling sharply over the year.
Whatever your view of the police, those are striking figures for a city region of nearly three million people, and they did not happen by accident. What is interesting is the explanation behind them, because it says something about where the whole idea of “security” is heading, not just for the police, but for the businesses, schools, landlords and venues that make up the rest of a city.
The instinctive assumption is that safer streets mean more officers walking them. More boots, more patrols, more visible presence. And there has been some of that. But the larger part of the shift has been quieter and, frankly, less photogenic. Greater Manchester Police has spent the past few years becoming a more data-led organisation: concentrating effort on a relatively small number of hotspots identified through analysis rather than hunch, attending every reported burglary as a matter of policy, and leaning on technology to find people and patterns faster than a human team ever could on its own.
Some of that technology has been contentious, and reasonably so. The force’s live facial recognition vans, deployed for the first time in late 2025 in places like Piccadilly Gardens and Bolton town centre, sit at the sharp end of a genuine national argument about privacy and consent. The automatic number plate cameras quietly feeding investigations raise their own questions. These are debates a city should have out loud, and Manchester is having them. But strip away the controversy for a moment and a more basic principle is visible underneath, one that has nothing to do with surveillance vans: the difference between watching something happen and actually knowing about it in time to act.
That distinction is the thread that connects the police story to everyone else. Because the truth is that most organisations are still firmly on the wrong side of it.
Think about how a typical Manchester business, school or managed building has handled security for the past twenty years. There are cameras, usually a lot of them. There is probably a separate system for the doors, another for the alarm, a sign-in book or a tablet at reception, and an emergency plan written down somewhere that most staff have read once. Each part works in isolation. None of them really talk to each other. And almost all of it is designed to be looked at after the fact. When something goes wrong, the footage exists, but it has to be hunted down, frame by frame, long after the moment to do anything useful has passed. It is the private-sector version of the very problem the police have spent years trying to solve: plenty of information, scattered everywhere, none of it reaching the right person at the right time.
The reason this matters now is that the tools to close that gap have finally become good enough, and cheap enough, to be worth a serious organisation’s attention. A new generation of what the industry rather drily calls physical security platforms is built around the same logic the police have adopted: bring the separate pieces into one place, let software watch continuously for the things that actually warrant a human response, and raise the alarm while there is still time to do something rather than the morning after. The interesting development is the use of artificial intelligence not to record more, but to filter, so that a flagged event reaches a manager’s phone in seconds, and a search through hours of footage takes the time it takes to type a sentence rather than the time it takes to watch an afternoon.
Companies working in this space, Coram among them, tend to make a point of running on the cameras an organisation already owns rather than insisting everything be ripped out and replaced, which historically has been the single biggest reason these upgrades never happen. That is a small detail with large consequences. It means the shift from passive recording to active awareness is no longer a capital project reserved for the largest operators. It is something a school, a logistics site on the edge of the city, a late-licence venue in the Northern Quarter or a landlord managing a half-empty building mid-redevelopment can plausibly do.
None of this is a substitute for policing, and nobody serious is claiming it is. The point is subtler. What Greater Manchester’s improving figures really demonstrate is not that cameras catch criminals, but that the value of any security system lies in how quickly it turns observation into action. The police have learned that lesson at the scale of a city region. The technology to apply the same lesson at the scale of a single building is now sitting on the table, and the organisations that pick it up will spend a lot less time, as so many still do, reviewing the footage of a problem they could not stop.
It would be easy to read the past year’s numbers as simply good news and leave it there. The more useful reading is that they are a proof of concept. Safety, it turns out, is less about how much you watch and more about how fast you understand what you are seeing. That is a principle every organisation in the city can borrow, and increasingly, the ones paying attention already are.
