How we judge and rank

A judgment is useful only when the method is visible. Here is ours.

Where a verdict on security devices comes from

Every verdict here is assembled in the same four steps, and each of them leaves something on the page you can check: we read the specifications, the manuals and the standards that apply to the product, we weigh what long-term owners consistently report about it, and we name the source of every figure we publish.

That last rule is the one that does the work. A number with a source can be argued with; a number without one is decoration. Where a figure exists in a manual and contradicts the marketing, we quote the manual. Where no reliable figure exists at all, the page says so and leaves the field empty rather than filling it with something plausible.

What we weigh

We rank on the specifications that decide the job and the ones that decide how long it keeps doing it. In practice that means five things:

• The number that governs the product. The illuminator behind a night image rather than the resolution printed on the box, the assumed event count behind a quoted battery runtime, the gap tolerance behind a contact sensor, the aperture behind a parcel box's stated volume, the reference edge of a cylinder length. Where a maker publishes a headline figure without the one that qualifies it, we say so rather than repeating it.
• What decides whether it is still watching in five years. A battery that can be swapped rather than a camera that must come off a gable to charge, a doorbell that rings a chime as well as a phone, a privacy shutter you can see from the doorway. Every category has one detail that decides whether a product is lived with or switched off after a month, and that is the one we describe.
• What is not in the box, and what is not free. The subscription several cameras need before they record anything, the hub a sensor cannot work without, the memory card sold separately, the longer screws a strike plate should have been supplied with. This is weighted more heavily here than on most sites, deliberately: a recurring plan usually outgrows the price difference between two models within a year.
• Batteries and parts, and whether they are available. What cards, drives, lithium cells and mounts cost over a few years; whether a battery is replaceable or sealed; whether a proprietary solar connector will still be sold when the panel fails. This is why a product that performs slightly worse sometimes outranks one that performs better.
• Whether it is stocked where you are. Each reference is confirmed market by market, and the page says which markets carry a live offer. A camera nobody sells where you live is not a recommendation, whatever it records.

Where the evidence on security devices comes from

Three sources, weighted differently and always named. Manufacturer specifications and manuals, cited as such rather than repeated as fact — they are reliable for dimensions and unreliable for durability. Declared standards where they exist (a UL 1037 or UL 687 grade, an ANSI/BHMA grade, the EN 1303 or EN 1143-1 class a European maker declares, an IP or IK rating), always flagged as a figure to verify on the current listing rather than a permanent guarantee. And the weight of long-term owner reports, which is the only practical source for the question that matters most: what is this product like in year three, after the warranty, after an app update, and after the first winter?

Where a claim rests on one source rather than three, the page says so. Where we could not source a figure at all, the field is left empty and the gap is stated. We never fill one in by inference.

What the ordering means

Each product card shows a rating out of ten. It sums up how the product stands within its category against the criteria above and the published specifications — it is not a measurement, and it only holds there: an 8.1 among safes and an 8.1 among cameras say nothing about each other, because no single figure can compare a $600 wired camera system with a $15 leak sensor. What matters more than the number is the order: within a category, the sequence is deliberate, and the reasons for it are in the text.

Within a hub the order runs entry, mid, then top. A buyer arrives with a budget rather than with an ambition, and a page that opens on the dearest system answers a question nobody asked.

The biases in security devices we designed against

Affiliate sites drift towards expensive products, towards whatever is newest, and towards whatever converts. Our defense is structural rather than a promise: no ranking on this site has ever been reordered because of what a product pays, several pages recommend a $15 accessory instead of a new device, and no product has ever been placed in exchange for anything.

There is a bias specific to this subject, and it is worth naming. Fear sells security equipment. Burglary statistics make effective copy and terrible advice, and a page that opens by frightening you tends to close by recommending the most expensive thing on it. No page here opens that way. Where the honest answer is that a device changes very little — a dummy camera, a second indoor sensor, a louder siren — the page says so.

The one bias we cannot remove is availability. We cover what is genuinely purchasable across Amazon's US, UK, German, French, Italian and Spanish stores, which means good equipment sold only through locksmiths and security installers is under-represented here. Where that matters to the recommendation, the page says so.

Resistance grades are reported, never derived

A lock, a safe and an alarm are judged in public by grades, and those grades are the easiest thing on a listing to repeat carelessly. The rule that governs every page touching them: every resistance grade, IP and IK rating, decibel figure and battery runtime on this site is reported from the manufacturer's own documentation, with its source named, and never calculated or inferred by us. Where a figure is not published, the page says it is not published rather than offering a plausible one.

Where a maker states a safe to UL 1037, a cylinder to EN 1303 or an alarm to a VdS class, we report that the manufacturer states it and link to what they publish. We are not a testing body and we do not assess conformity, so nothing here should be read as a judgment that a product meets a standard — and no product on this site is presented as approved by an insurer. Whether a given lock or safe satisfies your policy is a question for your policy, and the answer is not on a product page.

What these products actually do

One sentence governs the whole site, and it is the reason several pages here read less enthusiastically than their competitors: a lock delays, a camera records, an alarm alerts, and a safe buys time. None of them prevents anything. We use those verbs deliberately, because the alternative — writing as though a device makes a house safe — sells more equipment and leaves a reader with a belief that is not true.

Where this site stops

Nothing here replaces the manufacturer's instructions. Where a job belongs to a qualified person — altering a fire door, wiring a gate release into a mains circuit, fitting a safe into a structural floor — the page says so instead of talking you through it, even though a page that stops earns nothing.

And where a question is legal rather than technical — what may be in frame, how long footage may be kept, what a neighbor is entitled to ask — the page describes the rule, links to the authority that publishes it, and goes no further. Pages that touch that boundary say it in the text rather than in a disclaimer at the bottom.