Every automated system produces flags that are wrong, and the cost of acting on a wrong flag in a family is high. A false accusation about a child’s friend does not undo itself. So the grid exists to put a human between a pattern and a consequence.
It is also the layer that handles the things no software can: the neighbour who mentions a car that has been parked there three days, the school that calls, the relative whose account starts asking for money. A household’s real security has always run partly on people who pay attention. This layer gives that a structure.
It is also where the oldest kind of security lives. Address and presence de-listing, a hardened front door, a monitored response when something at the property is wrong, a movement plan for the households that genuinely need one — none of it is software, and all of it is the same job: fewer open doors, and someone watching the ones that remain.
What it does
- Human review before any flag becomes an action
- A named contact a family can reach, rather than a ticket queue
- Community and neighbourhood signal, gathered with consent
- Household briefings and drills, so the family recognises the play
- A written incident-response playbook, read before it is needed rather than during
- A review of the physical perimeter too — the doors, the grounds, and what makes a household less findable and better watched
The limit Human review is the slowest layer and the most expensive one. It is deliberately reserved for consequence — the machine handles volume, people handle decisions.