Our methodology

What we actually do

We are a research desk, not a test lab. For every product we consider, we pull the manufacturer's own specification sheet and, where it exists, the owner's manual — the primary sources for sensor size, resolution, autofocus system, video modes and record limits, stabilization, battery rating, ports, weight and dimensions. We cross-check those against the specifications published by other reputable outlets, and against the aggregated experience of owners. Then we do the arithmetic nobody else seems to: the real cost of a usable kit, the crop-factor equivalents, the cost of growing into a lens system.

What we do not do — and won't pretend to

We do not run a testing lab. We have not put these products through controlled measurements of dynamic range, autofocus hit-rate, rolling shutter or low-light performance, because doing that honestly requires equipment and controlled conditions we do not have. You will never see the phrase "in our testing" on this site, because it would be false. Where a measured number matters and someone credible has measured it, we cite them by name and link the source — we don't launder their authority by presenting it as our own.

• Units we claim to have tested: 0.
• Sponsored placements: 0.
• Manufacturers who see our copy before publication: 0.

Those are not gaps we're embarrassed about — they're claims we refuse to make. The alternative, which is most of this industry, is to assert testing that never happened.

The scoring rubric

Every product carries a score out of 10, and that score is the mean of a handful of named metrics that are visible on the product card — so you can see what it's made of and disagree with any single line. The exact metrics vary by category (a lens is judged on sharpness and aperture; a microphone on pickup pattern and background rejection), but they always follow the same logic:

• Core performance — the specs that decide the result: sensor and autofocus for cameras, aperture and sharpness for lenses, sustained write speed for SD cards.
• Fit for the use — how well it does the specific job the page is about (vlogging, portraits, travel), which is why the same camera can win one roundup and place mid-pack in another.
• Build & handling — ergonomics, weight, reliability as reported by owners.
• Value — performance per dollar at the live price, which is why cheaper picks routinely out-score flagships.

A score is a judgment from documented research, not a lab measurement. We think a transparent, reproducible judgment is worth more than an opaque one dressed up as science.

How prices work

Prices are never typed by hand. They come from Amazon's product data API and are stamped with the date they were fetched. If our price data is more than 48 hours old, the number is hidden and the button falls back to "Check price on Amazon" — the site would rather show you nothing than a figure that has rotted. Amazon's price at checkout is always the one that counts.

Product currency

We only recommend gear that is current and available in the United States. When a model is discontinued or restricted, it comes off the live picks — we don't leave dead links to make a list look full. Where a category leans heavily on one brand's availability, we deliberately include alternatives so a recommendation can't go dead overnight.

How often we update

Prices refresh continuously. Picks are reviewed when a successor model ships or a recommendation is discontinued, and each roundup carries a visible "Updated" date. If you spot something out of date or wrong, tell us via the contact form — corrections to factual errors go up within 48 hours with a note.