Research · 3 min read
You Have Cameras. Do They Cover the Places That Actually Matter?
Camera count is a poor proxy for usable coverage; priority points and failure causes lead to better fixes.
Twenty cameras, a clean grid of feeds, a monitoring contract signed — none of it answers the only question that matters: can you see the places trouble would actually come from? Our assessments name five such places before looking at a single camera. Across 30 consecutive properties, 25 had at least one of the five blind.
Our analysis of 150 priority-point observations across 30 properties. Scope: 30 consecutive property assessments, five predefined priority points each.
Key takeaways
- 25 of 30 assessed properties (83.3%) had at least one of their five priority points rated blind.
- 40 of 150 priority points (26.7%) fell outside usable coverage; the median property was blind on one of its five.
- 62.5% of blind points came from field-of-view (16) or fixed-obstruction (9) causes, which call for aiming or layout changes before new hardware.
- Audit the priority views in daylight and darkness before adding cameras, and re-run the audit after layout, lighting, or landscaping changes.
Should a property add cameras before auditing the views it already has?
Across 150 priority points in 30 consecutive property assessments, 25 properties had at least one blind priority point and 62.5% of blind points were attributed to field-of-view or fixed-obstruction problems. Audit the views before adding hardware.
The numbers behind the answer
Selected measures only. Denominators and interpretation stay attached so the headline cannot stand alone.
83.3%
Properties with a blind priority point
25 of 30 assessed properties had at least one of five priority points rated blind.
26.7%
Blind priority points
40 of 150 assessed priority points were outside usable coverage.
62.5%
Field-of-view or fixed-obstruction causes
25 of 40 blind points were attributed to those two causes.
A property can have cameras and still miss a priority view
The audit starts before any camera talk: name five priority points — the entrances, the loading area, the approach paths, the spots an incident would have to cross — then check whether each has a usable view. Across 30 properties, 25 had at least one blind point. The median property was blind on one of its five.
That’s why camera count is a weak way to buy security. Ten cameras aimed at the wrong places cover less than six aimed at the right ones. Coverage is a property of aiming, obstruction, light, and uptime — not of the invoice.
First action
Audit the views you need before deciding how much hardware to add.
Most blind points call for diagnosis—not an automatic camera purchase
The 40 blind points failed for six different reasons, and the mix is the useful part. Outside the field of view: 16. Fixed obstruction: 9. Insufficient light: 7. Glare: 5. Resolution too low at distance: 2. Camera simply offline: 1.
Each cause has its own fix, and most aren’t “buy another camera.” Field of view is an aiming or mounting job. Obstruction is a layout or overlap change. Light and glare need day-and-night tests. Resolution needs an identification test at real distance. Offline is maintenance and health monitoring. Diagnosis first, hardware second.
Evidence visual
Observed cause among 40 blind priority points
Shares use blind points—not all assessed points—as the denominator. Cause labels describe the primary recorded reason.
Run the audit in the conditions the view has to handle
The audit is repeatable, and the conditions it runs in matter as much as the checklist. A view that works at noon can be useless at 2 a.m. Define the point and the visual task — detection, recognition, identification — then test framing, obstructions, light, glare, resolution at distance, and system health, in daylight and darkness where both matter.
What this aggregate can’t tell you: whether fixing a blind point prevents losses (no incident data was joined to the gaps), or whether night blindness exceeds day blindness in general (that needs its own denominators). What it can tell you is where to look first. Re-run it after any layout, lighting, or landscaping change.
- Name each priority point and required visual task.
- Check frame coverage and likely path of travel.
- Check fixed and changing obstructions.
- Test daylight, low light, glare, and weather-relevant conditions.
- Verify resolution at actual distance and monitor uptime.
- Reassess after layout, lighting, landscaping, or operational changes.
40 of 150 priority points were blind, and two causes explain 62.5%
The audit puts a number on the geometry. Across 30 properties and five priority points each, 40 of 150 points were blind — 26.7% — and 25 of 30 properties (83.3%) had at least one. The causes concentrate: outside the field of view 16, fixed obstruction 9, insufficient light 7, glare 5, resolution 2, offline 1. Field-of-view and obstruction alone are 25 of 40, or 62.5%.
Those are the expected failures of camera geometry, and line-of-sight methods for mapping camera coverage formalize the same point: a camera covers a cone, obstructions cut it, and priority points often fall outside. Adding cameras without a coverage map adds cones, not coverage — which is why aiming and layout come before hardware.
The wider evidence keeps expectations calibrated. The 40-year systematic review of CCTV finds modest, offense-specific effects concentrated in certain settings, and situational crime prevention holds that removing opportunity matters more than adding hardware. A camera aimed at a priority point changes opportunity; one aimed elsewhere does not. That is where the coverage map meets the physical audit.
Order of operations
Map the views first; hardware only replaces coverage that geometry cannot fix.
Questions property teams ask
What counts as a functional blind spot?
A priority point is blind when it has no view or the available view cannot perform the predefined visual task. The exact rubric should be documented for each assessment.
Is a blind spot always fixed by adding another camera?
No. Aiming, mounting, obstruction, lighting, glare, resolution, or uptime can be the primary issue; each requires a different remedy.
How often should camera views be reassessed?
Reassess after layout, lighting, landscaping, access, or equipment changes and on a regular maintenance cadence appropriate to the property.
Does one blind point mean the whole camera system is ineffective?
No. It identifies one unmet priority view. The operational importance depends on the point and the task the system must support there.
Why do field-of-view problems dominate blind points?
Because camera coverage is a geometric cone. Aim, mounting height, and obstructions determine what falls inside it, so the most common failure is a priority point simply outside the view.
Our methods, limits, and sources
How we calculated this
This is original 911 Sentinel research — we gathered the records, ran every calculation below, and published the aggregate dataset.
We recorded five predefined priority points per property across 30 consecutive assessments, rated whether each has usable coverage, and assigned a primary cause to blind points.
- We validated property, priority-point, context, visibility, cause, and review fields.
- We calculated point-level and equal-property-weighted blind shares.
- We counted properties with at least one blind point and the distribution of blind counts.
- We ranked causes and preserved denominators for property, point-type, and context cross-tabs.
What this analysis cannot establish
- The aggregate covers 30 properties and five predefined priority points per property.
- Site selection, property mix, dates, and the exact usable-view rubric affect generalization.
- Usable-view ratings are made by the assessing responder; this aggregate reports point-level outcomes, not independent inter-rater agreement.
- No incident-outcome join supports a prevention or loss-reduction claim.
- Context counts are not context-specific blind rates without denominators.
Sources
The raw records come from the sources below; the study design, analysis, charts, and conclusions are our own.
- Kenneth Lewenhagen, Martin Boldt & Anton Borg (IEEE ISCC 2024) — Automated Generation of CCTV Camera Coverage Areas for Smart Cities Using Line-of-Sight Analysis
- Eric L. Piza, Brandon C. Welsh, David P. Farrington & Amanda L. Thomas (Criminology & Public Policy) — CCTV surveillance for crime prevention: A 40-year systematic review with meta-analysis
- Ronald V. Clarke (Crime and Justice) — Situational Crime Prevention
Related questions and practical guides
No obligation · free property walk
Audit the priority views before expanding the system
A site assessment can define the important views, test the existing system in relevant conditions, and separate adjustment, maintenance, lighting, and hardware needs.